一种利用浮标站资料改进海浪模式有效波高预报的方法-海洋预报

一种利用浮标站资料改进海浪模式有效波高预报的方法

单位：浙江省舟山市气象局, 浙江舟山 316021

分类号：P731.34

出版年·卷·期（页码）：2020·37·第一期（50-54）

摘要：

基于浮标站海浪历史数据，利用回归分析方法建立了海浪数值模式有效波高预报产品的一元二次回归方程订正统计模型。通过2017年7月1日—2018年10月10日期间业务试运行结果发现：订正方程能有效改善有效波高数值预报产品的预报精度，且预报时效越短订正效果越显著。其中，第6～11 h预报时效内的订正前后平均绝对误差值减小0.17～0.241 m，第6～18 h预报时效内订正前后均方根误差减小幅度为0.103～0.28 m。这说明应用订正统计模型对海浪模式输出产品进行订正，也是改进海浪模式预报准确率的一种有效途径。

Based on the historical wave data of the buoy station, this paper establishes a modified statistical model of the unary quadratic regression equation for the significant wave height prediction product of the wave numerical model using the regression analysis method. The results of the trial operation between July 1, 2017 and October 10, 2018 show that the revised equation has higher forecasting ability and can effectively improve the prediction accuracy of the significant wave height. The shorter prediction time, the more significant is the correction effect. The average absolute error decreases by 0.17-0.241 m and 0.103-0.28 m after correction within 6-11 h and 6-18 h forecast aging, respectively. It shows that establishing a modified statistical model to correct the output of wave model is an effective way to improve the accuracy of numerical wave prediction.

参考文献：

[1] Makarynskyy O. Improving wave predictions with artificial neural networks[J]. Ocean Engineering, 2004, 31(5-6):709-724.
[2] 许富祥. 海浪预报现状与未来[J]. 海洋预报, 2005, 22(S1):172-175.
[3] 孟雷, 闻斌, 姜洪峰, 等. 神经网络方法对海浪有效波高数值模拟的改进[J]. 海洋预报, 2010, 27(2):8-14.
[4] 毛科峰, 萧中乐, 王亮, 等. 数值模式与统计模型相耦合的近岸海浪预报方法[J]. 海洋学报, 2014, 36(9):18-29.
[5] Sverdrup H U, Munk W H. Wind, sea and swell:theory of relation for forecasting[M]. Washington:U. S. Navy Hydrographic Office, 1947:44.
[6] Pierson W J Jr, Neumann J G, James R W. Practical methods for observing and forecasting Ocean waves by means of wave spectra and statistics[M]. Washington:U. S. Navy Hydrographic Office, 1955:284.
[7] 中华人民共和国交通部. 海港水文规范[M]. 北京:人民交通出版社, 1998.
[8] Özger M. Significant wave height forecasting using wavelet fuzzy logic approach[J]. Ocean Engineering, 2010, 37(16):1443-1451.
[9] Tseng C M, Jan C D, Wang J S, et al. Application of artificial neural networks in typhoon surge forecasting[J]. Ocean Engineering, 2007, 34(11-12):1757-1768.
[10] Zamani A, Solomatine D, Azimian A, et al. Learning from data for wind-wave forecasting[J]. Ocean Engineering, 2008, 35(10):953-962.
[11] 梅婵娟, 赵栋梁, 史剑. 两种海浪模式对中国黄海海域浪高模拟能力的比较[J]. 海洋预报, 2008, 25(2):92-98.
[12] 江丽芳, 张志旭, 齐义泉, 等. WAVEWATCH Ⅲ和SWAN模式在南海北部海域海浪模拟结果的对比分析[J]. 热带海洋学报, 2011, 30(5):27-37.
[13] 李燕, 黄振, 张俊峰, 等. WAVEWAT CH Ⅲ模式在渤海海浪预报的应用与检验[J]. 气象与环境学报, 2014, 30(1):23-29.
[14] 徐丽丽, 肖文军, 石少华, 等. WaveWatch Ⅲ和SWAN模型嵌套技术在业务化海浪预报系统中的应用及检验[J]. 海洋通报, 2015, 34(3):283-294.
[15] Tolman H L, Banner M L, Kaihatu J M. The NOPP operational wave model improvement project[J]. Ocean Modelling, 2013, 70:2-10.
[16] Dietrich J C, Zijlema M, Allier P E, et al. Limiters for spectral propagation velocities in SWAN[J]. Ocean Modelling, 2013, 70:85-102.
[17] Reikard G, Rogers W E. Forecasting ocean waves:comparing a physics-based model with statistical models[J]. Coastal Engineering, 2011, 58(5):409-416.
[18] 朱智慧, 曹庆, 徐杰. 神经网络方法在上海沿海海浪预报中的应用[J]. 海洋预报, 2018, 35(5):25-33.

服务与反馈：

【文章下载】【发表评论】【查看评论】【加入收藏】