Neural network for wave forecasting among multi-stations

Unlike in the open sea, the use of wind information for forecasting waves may encounter more ambiguous uncertainties in the coastal or harbor area due to the influence of complicated geometric configurations. Thus this paper attempts to forecast the waves based on learning the characteristics of observed waves, rather than the use of the wind information. This is reported in this paper by the application of the artificial neural network (ANN), in which the back-propagation algorithm is employed in the learning process for obtaining the desired results. This model evaluated the interconnection weights among multi-stations based on the previous short-term data, from which a time series of waves at a station can be generated for forecasting or data supplement based on using the neighbor stations data. Field data are used for testing the applicability of the ANN model. The results show that the ANN model performs well for both wave forecasting and data supplement when using a short-term observed wave data.     

Wave hindcasting by coupling numerical model and artificial neural networks

By coupling numerical wave model (NWM) and artificial neural networks (ANNs), a new procedure for wave prediction is proposed. In many situations, numerical wave modeling is not justified due to economical consideration. Although incorporation of an ANN model is inexpensive, such a model needs a long time period of wave data for training, which is generally inconvenient to achieve. A proper combination of these two methods could carry the potentials of both. Based on the proposed approach, wave data are generated by a NWM by means of a short period of assumed winds at a concerned point. Then, an ANN is designed and trained using the above-mentioned generated wind-wave data. This ANN model is capable of mapping wind-velocity time series to wave height and period time series with low cost and acceptable accuracy. The method was applied for wave hindcasting to two different sites; Lake Superior and the Pacific Ocean. Simulation results show the superiority of the proposed approach.     

Wave model predictions in the Black Sea: Sensitivity to wind fields

This paper evaluates the impact of using different wind field products on the performance of the third generation wave model SWAN in the Black Sea and its capability for predicting both normal and extreme wave conditions during 1996. Wind data were obtained from NCEP CFSR, NASA MERRA, JRA-25, ECMWF Operational, ECMWF ERA40, and ECMWF ERA-Interim. Wave data were obtained in 1996 at three locations in the Black Sea within the NATO TU-WAVES project. The quality of wind fields was assessed by comparing them with satellite data. These wind data were used as forcing fields for the generation of wind waves. Time series of predicted significant wave height (H_mo), mean wave period (T_m02), and mean wave direction (DIR) were compared with observations at three offshore buoys in the Black Sea and its performance was quantified in terms of statistical parameters. In addition, wave model performance in terms of significant wave height was also assessed by comparing them against satellite data.The main scope of this work is the impact of the different available wind field products on the wave hindcast performance. In addition, the sensitivity of wave model forecasts due to variations in spatial and temporal resolutions of the wind field products was investigated. Finally, the impact of using various wind field products on predicting extreme wave events was analyzed by focussing on storm peaks and on an individual storm event in October 1996. The numerical results revealed that the CFSR winds are more suitable in comparison with the others for modelling both normal and extreme events in the Black Sea. The results also show that wave model output is critically sensitive to the choice of the wind field product, such that the quality of the wind fields is reflected in the quality of the wave predictions. A finer wind spatial resolution leads to an improvement of the wave model predictions, while a finer temporal resolution in the wind fields generally does not significantly improve agreement between observed and simulated wave data.     

Wave height forecasting in Dayyer, the Persian Gulf

Forecasting of wave parameters is necessary for many marine and coastal operations. Different forecasting methodologies have been developed using the wind and wave characteristics. In this paper, artificial neural network (ANN) as a robust data learning method is used to forecast the wave height for the next 3, 6, 12 and 24 h in the Persian Gulf. To determine the effective parameters, different models with various combinations of input parameters were considered. Parameters such as wind speed, direction and wave height of the previous 3 h, were found to be the best inputs. Furthermore, using the difference between wave and wind directions showed better performance. The results also indicated that if only the wind parameters are used as model inputs the accuracy of the forecasting increases as the time horizon increases up to 6 h. This can be due to the lower influence of previous wave heights on larger lead time forecasting and the existing lag between the wind and wave growth. It was also found that in short lead times, the forecasted wave heights primarily depend on the previous wave heights, while in larger lead times there is a greater dependence on previous wind speeds.     

Local model approximation in the real time wave forecasting

The notion of data assimilation is common in most wave predictions. This typically means nudging of wave observations into numerical predictions so as to drive the predictions towards the observations. In this approach, the predicted wave climate is corrected at each time of the observation. However, the corrections would diminish soon in the absence of future observations. To drive the model state predictions towards real time climatology, the updating has to be carried out in the forecasting horizon too. This could be achieved if the wave forecasting at the observational network is made available. The present study addresses a wave forecasting technique for a discrete observation station using local models. Embedding theorem based on the time-lagged embedded vector is the basis for the local model. It is a powerful tool for time series forecasting. The efficiency of the forecasting model as an error correction tool (by combining the model predictions with the measurements) has been brought up in a forecasting horizon from few hours to 24 h. The parameters driving the local model are optimised using evolutionary algorithms.     

Significant wave height forecasting using wavelet fuzzy logic approach

Wave heights and periods are the significant inputs for coastal and ocean engineering applications. These applications may require to obtain information about the sea conditions in advance. This study aims to propose a forecasting scheme that enables to make forecasts up to 48 h lead time. The combination of wavelet and fuzzy logic approaches was employed as a forecasting methodology. Wavelet technique was used to separate time series into its spectral bands. Subsequently, these spectral bands were estimated individually by fuzzy logic approach. This combination of techniques is called wavelet fuzzy logic (WFL) approach. In addition to WFL method, fuzzy logic (FL), artificial neural networks (ANN), and autoregressive moving average (ARMA) methods were employed to the same data set for comparison purposes. It is seen that WFL outperforms those methods in all cases. The superiority of the WFL in model performances becomes very clear especially in higher lead times such as 48 h. Significant wave height and average wave period series obtained from buoys located off west coast of US were used to train and test the proposed models.     

Simulation of the extreme waves generated by typhoon Bolaven (1215) in the East China Sea and Yellow Sea

Record-breaking high waves occurred during the passage of the typhoon Bolaven (1215) (TYB) in the East China Sea (ECS) and Yellow Sea (YS) although its intensity did not reach the level of a super typhoon. Winds and directional wave measurements were made using a range of in-situ instruments mounted on an ocean tower and buoys. In order to understand how such high waves with long duration occurred, analyses have been made through measurement and numerical simulations. TYB winds were generated using the TC96 typhoon wind model with the best track data calibrated with the measurements. And then the wind fields were blended with the reanalyzed synoptic-scale wind fields for a wave model. Wave fields were simulated using WAM4.5 with adjustment of C_d for gust of winds and bottom friction for the study area. Thus the accuracy of simulations is considerably enhanced, and the computed results are also in better agreement with measured data than before. It is found that the extremely high waves evolved as a result of the superposition of distant large swells and high wind seas generated by strong winds from the front/right quadrant of the typhoon track. As the typhoon moved at a speed a little slower than the dominant wave group velocity in a consistent direction for two days, the wave growth was significantly enhanced by strong wind input in an extended fetch and non-linear interaction.     

Effect of climate change on design wind at the Indian offshore locations

The impact of climate change on design wind speeds corresponding to different return periods at two selected offshore locations in India has been assessed. Extreme daily wind speeds corresponding to various return periods were derived based on the observations made by wave rider buoys during the period 1998–2005. Thereafter, the future climate over the next century was simulated at these locations using the input from the climate model: GCM-CGCM3 corresponding to the A2 scenario. The underlying downscaling model was developed with the help of artificial neural networks and using observed wind as output. The local wind speeds corresponding to these projected wind data were generated for the next century and return period wind speeds were extracted by the distribution fitting. Comparison of design wind speeds derived with and without consideration of future climate showed that the magnitude of the long term wind speed would certainly and significantly increases if the effect of global climate change is incorporated in the analysis. For the two locations considered, the increase in the 100-year wind was found to be varying from 44% to 74%.     

Uncertainty Analysis of Wind-Wave Predictions in Lake Michigan

With all the improvement in wave and hydrodynamics numerical models, the question rises in our mind that how the accuracy of the forcing functions and their input can affect the results. In this paper, a commonly used numerical third-generation wave model, SWAN is applied to predict waves in Lake Michigan. Wind data are analyzed to determine wind variation frequency over Lake Michigan. Wave predictions uncertainty due to wind local effects are compared during a period where wind has a fairly constant speed and direction over the northern and southern basins. The study shows that despite model calibration in Lake Michigan area, the model deficiency arises from ignoring wind effects in small scales. Wave prediction also emphasizes that small scale turbulence in meteorological forces can increase prediction errors by 38%. Wave frequency and coherence analysis show that both models can predict the wave variation time scale with the same accuracy. Insufficient number of meteorological stations can result in neglecting local wind effects and discrepancies in current predictions. The uncertainty of wave numerical models due to input uncertainties and model principals should be taken into account for design risk factors.     

9914号(Dan)台风浪的后报试验研究

利用WAM第三代海浪模式的第四版本(WAMC4)对40a来造成福建沿海灾害最严重的9914号台风海浪过程进行了后报试验，并与近岸常规观测和卫星高度计有效波高资料进行了比较。与常规观测站的比较结果表明，WAMC4能较好地再现海浪的发展过程。后报结果与TOPEX/POSEIDON和ERS-2卫星观测资料的对比研究表明，风速的后报结果与卫星观测有较好的一致性，但海浪的后报比卫星高度计反演的有效波高整体略偏低。     

非结构网格模型在闽东沿海台风浪模拟中的应用

基于加密的非结构三角网格,以Holland模型风场叠加美国国家环境预报中心（NCEP）海面风场构造的合成风场驱动第三代浅水波浪数值模型（SWAN）对2017年影响闽东海域的“纳沙”和“泰利”台风过程进行数值模拟,并运用浮标站的实测数据对模拟结果进行验证.结果表明,模型计算的风速、有效波高与实测值符合较好,合成风场能较好地模拟台风期间的风速变化过程,SWAN模式能够合理地再现闽东沿海台风浪的时空分布特征.由模拟结果可见：台风“纳沙”中心越过台湾岛进入台湾海峡北部海面,受海峡地形的约束,其波浪场呈NE—SW向椭圆状分布,北部海域的浪高大于南部,闽东沿海遍布大范围的巨浪到狂浪;超强台风“泰利”未登陆闽东,当其台风中心与大陆的距离最近时,海面波浪场分布与台风风场结构一致,台风中心附近海域为14 m以上的怒涛区,巨浪遍布于闽东沿海.研究结果可为闽东沿海台风浪灾害预警和应急管理提供技术支撑和参考依据.     

New Interpretation of Dependence of Wind Stress on Wave State

Based on observations from buoys, it is found that the wave age is well correlated with the nondimensional wave height, and this correlation is best described by a 3/5-power law. This similarity law is valid in the cases of wind waves as well as swells under natural sea states. On the basis of the 3/5-power law combined with the well-known 3/2-power law, it is shown that the wave-induced wind stress increases rapidly with wave age, indicating that the traditional observations or analytic techniques have only given the turbulent Reynolds stress induced by short wind waves, but excluded the long-wave-induced wind stress. The latter constitutes a small fraction to the total wind stress when the wave age is smaller than 1.0. The increase of sea-surface roughness with wave age can be attributed to wave breaking.     

Soft computing approach for real-time estimation of missing wave heights

This paper presents soft computing approach for estimation of missing wave heights at a particular location on a real-time basis using wave heights at other locations. Six such buoy networks are developed in Eastern Gulf of Mexico using soft computing techniques of Artificial Neural Networks (ANN) and Genetic Programming (GP). Wave heights at five stations are used to estimate wave height at the sixth station. Though ANN is now an established tool in time series analysis, use of GP in the field of time series forecasting/analysis particularly in the area of Ocean Engineering is relatively new and needs to be explored further. Both ANN and GP approach perform well in terms of accuracy of estimation as evident from values of various statistical parameters employed. The GP models work better in case of extreme events. Results of both approaches are also compared with the performance of large-scale continuous wave modeling/forecasting system WAVEWATCH III. The models are also applied on real time basis for 3 months in the year 2007. A software is developed using evolved GP codes (C++) as back end with Visual Basic as the Front End tool for real-time application of wave estimation model.     

Derivation of coastal wind and wave parameters from offshore measurements of TOPEX satellite using ANN

The paper discusses an artificial neural network (ANN) approach to project information on wind speed and waves collected by the TOPEX satellite at deeper locations to a specified coastal site. The observations of significant wave heights, average wave period and wind speed at a number of locations over a satellite track parallel to a coastline are used to estimate corresponding values of these three parameters at the coastal site of interest. A combined network involving an input and output of all the three parameters, viz., wave height, period and wind speed instead of separate networks for each one of these variables was found to be necessary in order to train the network with sufficient flexibility. It was also found that network training based on statistical homogeneity of data sets is essential to obtain accurate results. The problem of modeling wind speeds that are always associated with very high variations in their magnitudes was tackled in this study by imparting training in an innovated manner.     

A forecasting model for wave heights based on a long short-term memory neural network

To explore new operational forecasting methods of waves, a forecasting model for wave heights at three stations in the Bohai Sea has been developed. This model is based on long short-term memory(LSTM) neural network with sea surface wind and wave heights as training samples. The prediction performance of the model is evaluated,and the error analysis shows that when using the same set of numerically predicted sea surface wind as input, the prediction error produced by the proposed LSTM model at S...     

Wind-generated waves in Hurricane Juan

We present numerical simulations of the ocean surface waves generated by hurricane Juan in 2003 as it reached its mature stage (travelling from deep waters off Bermuda to Nova Scotia and making landfall near Halifax) using SWAN (v.40.31) nested within WAVEWATCH-III (v.2.22; denoted WW3) wave models, implemented on multiple-nested domains. As for all storm-wave simulations, spectral wave development is highly dependent on accurate simulations of storm winds during its life cycle. Due to Juan’s rapid translation speed (accelerating from 2.28 m s⁻¹ on 27 September, 1200 UTC to 20 m s⁻¹ on 29 September, 1200 UTC), an interpolation method is developed to blend observed hurricane winds with numerical weather prediction (NWP) model winds accurately. Wave model results are compared to in situ surface buoys and ADCP wave data along Juan’s track. At landfall, Juan’s maximum waves are mainly swell-dominated and peak waves lag the occurrence of the maximum winds. We explore the influence of surface waves on the wind and show that the accuracy of the wave simulation is enhanced by introducing swell and Stokes drift feedback mechanisms to modify the winds, and by limiting the peak drag coefficient under high wind conditions, in accordance with recent theoretical and experimental results.     

唐山近海海域波浪要素特征分析

本文基于唐山近海海域1#、2#浮标2017年4月至11 月实时海浪观测数据及部分风速风向数据, 对唐山近海海域波浪有效波高、有效波向、有效波周期等波参数特征进行了统计分析, 并利用origin 软件对波参数与风速、风向相关性进行了研究。研究结果表明： 1#、2# 浮标海域常浪向为SSW、SW、SSE, 常浪向有效波高均以0.2 ～ 0.4 m 小浪及3 ～ 4 s 短周期为主,有效波高1 m 以上较大波浪极少出现; 该海域波浪以风浪为主, 波浪破碎速度较快, 有效波高与风速相关性较强, 相关系数r 为0.71, 风向与波向、有效波高与周期基本无相关性, 该研究资料可为海上活动及防灾减灾提供技术依据。     

台风浪模拟预报中的风场比较研究

在对模拟台风浪时海浪模式常用的经验模型风场和多重嵌套中尺度气象数值模式风场的结构和时间演变特征进行对比分析的基础上,分别采用这两种风场资料,应用最新版本的第三代海浪模式SWAN对Winnie(1997)引起的台风浪进行了模拟,将模拟的有效波高与TOPEX/POSEIDON和ERS-2卫星高度计资料作了详细的对比分析。结果表明,经验模型风场对实际台风风场的刻画存在诸多缺陷,这些缺陷对于台风浪的准确模拟产生了不可忽视的影响,采用模式风场试验的模拟效果优于采用模型风场的试验。论文提出了在运用海浪模式模拟台风浪时用数值模式模拟风场替代经验模型风场的必要性。     

Has wind–wave modeling reached its limit?

This article uses a comparison of four different numerical wave prediction models for hindcast wave conditions in Lake Michigan during a 10-day episode in October 1988 to illustrate that typical wave prediction models based on the concept of a wave energy spectrum may have reached a limit in the accuracy with which they can simulate realistic wave generation and growth conditions. In the hindcast study we compared the model results to observed wave height and period measurements from two deep water NOAA/NDBC weather buoys and from a nearshore Waverider buoy. Hourly wind fields interpolated from a large number of coastal and overlake observations were used to drive the models. The same numerical grid was used for all the models. The results show that while the individual model predictions deviate from the measurements by various amounts, they all tend to reflect the general trend and patterns of the wave measurements. The differences between the model results are often similar in magnitude to differences between model results and observations. Although the four models tested represent a wide range of sophistication in their treatment of wave growth dynamics, they are all based on the assumption that the sea state can be represented by a wave energy spectrum. Because there are more similarities among the model results than significant differences, we believe that this assumption may be the limiting factor for substantial improvements in wave modeling.     

Prediction of extreme wave heights using neuro wavelet technique

Precise prediction of extreme wave heights is still an evading problem whether it is done using physics based modeling or by extensively used data driven technique of Artificial Neural Network (ANN). In the present paper, Neuro Wavelet Technique (NWT) is used specifically to explore the possibility of prediction of extreme events for five major hurricanes Katrina 2005, Dean 2007, Gustav 2008, Ike 2008, Irene 2011 at four locations (NDBC wave buoys stations)¹ namely; 42040, 42039, 41004, 41041 in the Gulf of Mexico. Neuro Wavelet Technique is employed by combining Discrete Wavelet Transform and Artificial Neural Networks. Discrete wavelet transform analyzes frequency of signal with respect to time at different scales. It decomposes time series into low (approximate) and high (detail) frequency components. The decomposition of approximate components (extreme events in the ocean wave series) can be carried out up to the desired multiple levels in order to provide relatively smooth varying amplitude series. This feature of wavelet transforms make it plausible for predicting extreme events with a better accuracy. In the present study third, fifth and seventh level of decompositions are used which facilitates 3 to 7 times filtering of low frequency events and seems to pay the dividend in the form of better prediction accuracy at extreme events. To develop these Neuro wavelet models to forecast the waves with lead times of 12 hr to 36 hr in advance, previously measured significant wave heights at same locations were used. The results were judged by wave plots, scatter plots and other error measures. From the results it can be concluded that the Neuro Wavelet Technique can be employed to solve the ever eluding problem of accurate forecasting of the extreme events.