西北太平洋热带气旋非对称大风风圈半径预测-海洋预报

西北太平洋热带气旋非对称大风风圈半径预测

单位：国家海洋局东海预报中心, 上海 200136

分类号：P444

出版年·卷·期（页码）：2021·38·第三期（68-75）

摘要：

利用中央气象台2014—2018年发布的西北太平洋热带气旋分析资料，建立神经网络模型对热带气旋7级、10级和12级非对称大风风圈半径（R7、R10和R12）进行预测研究，为热带气旋的尺度和风场预测提供基础。选取1510、1521、1718、1807、1808和1822号台风对模型进行后报检验，结果显示：模型能较好地计算出热带气旋非对称大风风圈半径，6 h后报的R7平均绝对误差介于15～40 km，R10误差在5～15 km，R12误差<10 km，误差随着后报时效的增加有所增大，各级大风风圈半径的预测平均相对误差较为接近，在5%～15%之间。基于该模型，除后报检验选用的TC外，增加1909号台风对大风风圈半径进行预测，R7、R10和R12最近时效的预测平均绝对误差分别为33 km、20 km和10 km，预测较后报误差有所增大，检验台风总体平均的平均相对误差在10%～20%之间，NE象限预测误差较其他象限偏大。神经网络模型可作为热带气旋非对称大风风圈半径预测的有效手段。

Using the TC analysis data in Northwest Pacific from 2014 to 2018 released by the National Meteorological Center, a neural network model is established to predict the asymmetrical gale force wind radii (R7, R10, R12) for 6 h, 12 h and 24 h, which provides basis for the predictions of TC scale and wind field. The TC No. 1510, No. 1521, No. 1718, No. 1807, No. 1808 and No.1822, which have significant impacts on China in recent years, are used for the hindcast skill assessment of the model. The results show that the neural network model can reasonably depict the asymmetrical characteristics of the tropical cyclones. The 6 h-hindcast mean absolute errors (MAEs) of R7, R10 and R12 are 15～40 km, 5～15 km and less than 10 km, respectively, increasing with hindcasting period. The forecast mean relative errors (MREs) of different gale force wind radii are close, ranging from 5% to 15%. Taking TC No. 1909 into account, the model is employed to forecast the gale force wind radii for 7 typhoon processes. The MAEs of R7, R10 and R12 for the nearest forecast period are 33 km, 20 km and 10 km, respectively. The forecast MREs are between 10% and 20%, and the MAE in NE quadrant is greater than that in the other quadrants. It turns out that the neural network model is effective and could be used as an operational model for tropical cyclone gale force wind radii forecast.

参考文献：

[1] Knaff J A, Sampson C R, DeMaria M. An operational statistical typhoon intensity prediction scheme for the Western North Pacific[J]. Weather and Forecasting, 2005, 20(4):688-699.
[2] Judt F, Chen S S. Tropical cyclone intensity forecasting:predictability and uncertainty[C]//Proceedings of the 26th Conference on Weather Analysis and Forecasting/22nd Conference on Numerical Weather Prediction. Atlanta:The Georgia World Congress Center, 2014.
[3] Zhang Y J, Meng Z Y, Zhang F Q, et al. Predictability of tropical cyclone intensity evaluated through 5-yr forecasts with a convection-permitting regional-scale model in the Atlantic Basin[J]. Weather and Forecasting, 2014, 29(4):1003-1023.
[4] Goerss J S, Sampson C R, Gross J M. A history of western north pacific tropical cyclone track forecast skill[J]. Weather and Forecasting, 2004, 19(3):633-638.
[5] 于玉斌. 中国近海热带气旋强度突变研究[M]. 北京:气象出版社, 2012.
[6] 陈联寿, 孟智勇. 我国热带气旋研究十年进展[J]. 大气科学, 2001, 25(3):420-431.
[7] 端义宏, 余晖, 伍荣生. 热带气旋强度变化研究进展[J]. 气象学报, 2005, 63(5):636-645.
[8] 黄荣辉, 陈光华. 西北太平洋热带气旋移动路径的年际变化及其机理研究[J]. 气象学报, 2007, 65(5):683-694.
[9] Brand S. Very large and very small typhoons of the Western North Pacific ocean[J]. Journal of the Meteorological Society of Japan. Ser. II, 1972, 50(4):332-341.
[10] 雷小途, 陈联寿. 热带气旋风场模型构造及特征参数估算[J]. 地球物理学报, 2005, 48(1):25-31.
[11] 朱首贤, 沙文钰, 丁平兴, 等. 近岸非对称型台风风场模型[J]. 华东师范大学学报(自然科学版), 2002(3):66-71.
[12] Knaff J A, Sampson C R, DeMaria M, et al. Statistical tropical cyclone wind radii prediction using climatology and persistence[J]. Weather and Forecasting, 2007, 22(4):781-791.
[13] Sampson C R, Knaff J A. A consensus forecast for tropical cyclone gale wind radii[J]. Weather and Forecasting, 2015, 30(5):1397-1403.
[14] Fujita T. Pressure distribution in typhoon[J]. Geophysics Magazine, 1952, 23:437-452.
[15] Jelesnianski C P. A numerical calculation of storm tides induced by a tropical storm impinging on a continental shelf[J]. Monthly Weather Review, 1965, 93(6):343-358.
[16] Holland G J. An analytic model of the wind and pressure profiles in hurricanes[J]. Monthly Weather Review, 1980, 108(8):1212-1218.
[17] Baik J J, Paek J S. A neural network model for predicting typhoon intensity[J]. Journal of the Meteorological Society of Japan. Ser. II, 2000, 78(6):857-869.
[18] Lee T L. Neural network prediction of a storm surge[J]. Ocean Engineering, 2006, 33(3-4):483-494.
[19] 黄小刚, 费建芳, 陈佩燕. 利用神经网络方法建立热带气旋强度预报模型[J]. 应用气象学报, 2009, 20(6):699-705.
[20] Sampson C R, Fukada E M, Knaff J A, et al. Tropical cyclone gale wind radii estimates for the western North Pacific[J]. Weather and Forecasting, 2017, 32(3):1029-1040.
[21] 陈联寿, 刘式适. 关于热带气旋尺度变化机制的初步分析[J]. 热带气象学报, 1997, 13(2):105-111.
[22] 胡桂萍, 王咏青, 马红云. 西北太平洋热带气旋尺度的气候特征研究[J]. 山东气象, 2008, 28(3):1-8.

服务与反馈：

【文章下载】【发表评论】【查看评论】【加入收藏】