西北太平洋热带气旋频次的延伸期动力-统计预报方法和评估

介绍了西北太平洋热带气旋(TC)频次的延伸期预报方法,比较了新构建的动力-统计和统计预报模型的预测技巧,并探讨了预报误差来源及改进方向。动力-统计预报模型是基于动力模式预测的热带季节内振荡(ISO)信号及ISO-TC生成的同期统计关系来进行预报;统计预报模型则是基于TC生成的前兆ISO信号建模预报。预报评估结果显示,动力-统计混合预报模型的预报技巧高于统计预报模型,原因在于影响TC次季节变化的前兆信号并不稳定,且随着预报超前时间迅速消散,无法提供有效且稳定的可预报源;相反地,TC生成与同期的ISO背景场显著相关,动力模式对ISO(预报因子)有较好的预报能力,因此动力-统计相结合的预报方法为TC延伸期预报提供了有效途径。虽然目前动力-统计预报模型的预报技巧可达5～6周,但仍有进一步改进和提高的空间。通过对不同类型TC预报技巧检验和误差分析,研究认为年际和年代际背景场对ISO调控TC活动的影响不可忽略,且热带外ISO信号(如罗斯贝波破碎和西风急流强度等)对TC频次和轨迹也有显著影响,这些因子为TC延伸期预报提供了潜在可预报源。     

用数值预报释用方法做近海及登陆热带气旋强度预报

这是一个对GFS数值预报产品进行解释应用的方法.将支持向量机(SVM)回归方法应用于近海和登陆热带气旋(TC)的强度预报.从其本身强度,影响范围内气象因子情况,地形因子等三个方面,设计相关因子,建立预报模式,用来预报12、24、36、48、60和72 h的TC强度.总体上模式强度预报结果与中央气象台的预报结果相近,优于气候持续法的预报;趋势预报优势明显,可高出7～12个百分点.表明可以成为台风强度预报的另一个工具,投入业务应用.     

基于路径相似的登陆热带气旋降水之动力-统计集合预报模型

数值天气预报(NWP)过去几十年在热带气旋(TC)预报方面的最大进步是越来越准确的路径预报。对于登陆TC降水的预报,目前以数值模式为代表的技术手段预报能力还十分有限。围绕动力-统计结合之方法研究,初步发展了登陆热带气旋降水(LTP)预报的一种新方法:基于路径相似的登陆热带气旋降水之动力统计集合预报(LTP_DSEF)模型。该方法主要分为五步:TC路径预报、相似路径TC识别、其他特征相似性的判别、TC降水集合预报和最佳预报方案选择;涉及两个关键技术:TC降水分离的客观天气图分析法(OSAT)和TC路径相似面积指数(TSAI)。LTP DSEF模型对2012-2016年影响华南地区出现最大日降水量≥100 mm的21个TC的定量降水预报(QPF)试验结果显示,该模型对登陆TC过程降水的预报结果优于动力模式。登陆TC过程降水≥50 mm情况下,建模样本和独立样本平均TS评分均高于动力模式(EC、GFS、T639)相应的最好表现。对LTP_DSEF模型三个最佳方案的参数取值分析显示,起报时刻参数设定为最临近影响时刻即TC对陆地产生降水的前一天12:00 UTC、集合参数取最大值时预报效果稳定趋好。     

影响广西的热带气旋年频数的BP神经网络预测模型

对影响广西的热带气旋(TC)年频数与大气环流的关系进行分析表明,TC年频数与全球范围大气环流异常有密切关系,特别是春季南半球中高纬度环流异常和低纬越赤道气流异常.利用相关分析从春季全球大气环流场中选择初选预报因子,然后对初选预报因子作EOF展开构造综合预报因子,运用BP神经网络方法建立TC年频数预报模型,并对所建立的模型进行独立样本试验.结果表明,该预报模型对历史样本拟合精度高,试报效果优于传统的逐步回归模型,可在汛期预测业务中应用.     

基于前传式网络逼近的太平洋副热带高压活动的诊断预测

基于前传式神经网络BP算法(BackpropagationNeuralNetwork)和回归模型,探讨了西太平洋副高面积指数同赤道东太平洋海温及赤道纬向风之间非线性分类和映射逼近的建模方法和效果比较.结果表明,前传式网络,特别是回归网络预报模型具有较好拟合精度和预报效果及比较实用的预报时效.     

利用人工神经网络模型预测西北太平洋热带气旋生成频数

通过对60年(1950~2009年)北半球夏、秋季(6~10月)热带气旋(TC)频数与春季(3~5月)大尺度环境变量的相关分析,挑选出8个相关性较高的前期预报因子建立人工神经网络(ANN)模型,对2010~2017年8年夏、秋季TC频数进行回报,并将回报结果与传统多元线性回归(MLR)方法所得结果进行对比分析。结果表明,ANN模型对60年历史数据的拟合精度高,相关系数高达0.99,平均绝对误差低至0.77。在8年回报中,ANN模型相关系数为0.80,平均绝对误差为1.97;而MLR模型相关系数仅为0.46,平均绝对误差为3.30。ANN模型在历史数据拟合和回报中的表现都明显优于MLR模型,未来可考虑应用于实际的业务预测中。     

支持向量机(SVM)方法在气象预报中的个例试验

SVM(Support Vector Machine)方法的基本思想是:通过非线性映射把样本空间映射到一个高维乃至于无穷维的特征空间(Hilbert空间),在特征空间中寻求最优划分或回归线性超平面,从而解决样本空间中的高度非线性分类和回归等问题.利用1998～2000年9～11月T106-36小时预报的各种输出产品构造因子,以单站有无降水为预报对象,采用MOS法方式构造样本,通过训练建立了单站的晴雨SVM分类预报模型,利用1990～2000年4～9月ECMWF北半球的500hPa高度、850hPa温度、地面气压的0小时分析场资料,确定关键区域,构造预报因子,以PP法方式构造样本,通过训练建立了四川盆地内单站气温的SVM回归预报模型,并进行了模拟试验,结果表明,无论是单站晴雨的SVM分类预报模型还是单站平均气温的SVM回归预报模型都显示出了良好的预报能力.     

登陆热带气旋路径和强度预报的效益评估初步研究

近年来有关热带气旋(TC)灾情的评估指标和方法的研究取得明显进展,但较少涉及TC预报对减少灾害损失的贡献(即效益)分析。基于中央气象台的TC实时路径和强度预报,针对登陆中国大陆的TC,初步分析了TC的路径和强度预报误差与其造成的直接经济损失之间的可能关系,并在此基础上建立了包含TC路径和强度预报误差的TC直接经济损失的预估模型。TC登陆前后24 h的路径和强度预报误差与TC所致直接经济损失均呈正相关关系;对于单个登陆TC而言,若24 h TC路径预报误差每减小1 km可减少因灾直接经济损失约0.97亿元,若强度预报每减小1 m/s可减少因灾直接经济损失约3.8亿元(以2014年为基准年)。可见,提高TC路径和强度预报精度对于减灾的效益巨大,且当前尤以提高强度预报能力的效益为佳。      

基于数值预报产品的温度释用方法比较

为了选取适当的预报方法,运用于业务预报工作中的客观要素预报,我们以ECMWF数值预报输出产品为基础,对卡尔曼滤波(Klm)、最优子集回归(OSR)和岭回归(RR)三种方法针对温江、宜宾、内江、达县2005年6～7月的温度预报进行了探讨.结果表明:在样本完全相同的情况下,三种预报方法效果相当,但从运行速度来看,RR最快,Klm 其次,OSR 最慢;在预报方法相同而样本长度不同的情况下,样本长度为60天的预报效果较好.     

对流参数化影响热带气旋的模拟研究

利用GRAPES-TCM模式对2008年登陆我国的9个热带气旋(TC)进行了44次试验,分析了积云对流参数化方案Kain-Fritsch (KF)方案与Betts—Miller-Janjic(BMJ)方案对TC预报的影响.结果表明,KF方案预报TC的总体效果要好于BMJ方案,BMJ方案的优势主要体现在对强TC强度的预报.不同的对流参数化方案对TC路径的影响没有明显差异,但对TC强度和降水的影响与TC初始强度有关;不同的对流参数化方案预报的TC强度和降水强度各不相同,但不同方案预报TC强度的差异与TC降水强度的差异基本一致.采用不同的对流参数化方案预报TC强度和降水随着TC初始强度的不同而表现出不同的特点.     

主分量分析在热带气旋强度客观预报中的试用

为了解主分量因子分析在热带气旋强度客观预报中的应用效能,在NCEP再分析资料、T106L19模式产品和热带气旋历史观测资料基础上,采用主分量因子分析技术,结合多元线性回归和BP型人工神经网络,开展了西北太平洋热带气旋的强度客观预报技术研究试验。试验包含完全预报法原理下的预报因子有无主分量分析、线性与非线性预报方法建模等方面的比较。结果表明,预报因子的主分量分析通过降低线性回归和BP人工神经网络模型的维数,提高因子间独立性,可减小模型强度预报平均绝对误差,提高模型实际预报能力。     

Wilcoxon秩和检验在热带气旋强度预报方法评定中的应用

介绍了一种非参数检验—Wilcoxon秩和检验,并将其应用于热带气旋(tropical cyclone,TC)强度预报方法评定。由于在TC强度预报评定中,预报误差的分布是非正态分布,因此无法使用常用的参数检验,必须用非参数检验。结果表明,将Wilcoxon秩和检验应用于TC强度预报方法的评定,有利于更加深入理解TC强度预报方法,提高预报准确性。     

A NOVEL CLASSIFICATION METHOD FOR TROPICAL CYCLONE INTENSITY CHANGE ANALYSIS BASED ON HIERARCHICAL PARTICLE SWARM OPTIMIZATION ALGORITHM

Based on the tropical cyclone (TC) observations in the western North Pacific from 2000 to 2008, this paper adopts the particle swarm optimization (PSO) algorithm of evolutionary computation to optimize one comprehensive classification rule, and apply the optimized classification rule to the forecasting of TC intensity change. In the process of the optimization, the strategy of hierarchical pruning has been adopted in the PSO algorithm to narrow the search area, and thus to enhance the local search ability, i.e. hierarchical PSO algorithm. The TC intensity classification rule involves core attributes including 12-HMWS, MPI, and Rainrate which play vital roles in TC intensity change. The testing accuracy using the new mined rule by hierarchical PSO algorithm reaches 89.6%. The current study shows that the novel classification method for TC intensity change analysis based on hierarchic PSO algorithm is not only easy to explain the source of rule core attributes, but also has great potential to improve the forecasting of TC intensity change.     

2009—2015年ECMWF热带气旋集合预报的检验及分析

本文从四个方面检验分析了ECMWF 2009—2015年西北太平洋热带气旋集合平均预报性能。结果表明：集合预报对路径的预测能力逐年提高,对强度预报整体偏弱。随着热带气旋强度增强,集合预报对移速和移向的预测能力提高,而移向预报偏左、移速预报偏慢、强度预测偏弱的现象较明显。将影响热带气旋的引导气流分为偏强、中等、偏弱三类,引导气流偏弱时热带气旋移动偏慢,因此移向预报的不确定性大;而引导气流偏强时热带气旋移向明确,只是移速预报不稳定。进入南海的三类路径热带气旋,集合预报对西行、西北行两类的移速、移向预报效果较好,而西行后北折的预报较差,在热带气旋北折前,移向预报发散度很大,向北转折后移向趋于稳定,移速预报的误差相对较大。这几种情形的检验结果,在热带气旋集合预报的业务应用中值得注意。     

Experiments with the Improved Dynamical-Statistical-Analog Ensemble Forecast Model for Landfalling Typhoon Precipitation over South China

In recent work, three physical factors of the Dynamical-Statistical-Analog Ensemble Forecast Model for Landfalling Typhoon Precipitation (DSAEF_LTP model) have been introduced, namely, tropical cyclone (TC) track, TC landfall season, and TC intensity. In the present study, we set out to test the forecasting performance of the improved model with new similarity regions and ensemble forecast schemes added. Four experiments associated with the prediction of accumulated precipitation were conducted based on 47 landfalling TCs that occurred over South China during 2004-2018. The first experiment was designed as the DSAEF_LTP model with TC track, TC landfall season, and intensity (DSAEF_LTP-1). The other three experiments were based on the first experiment, but with new ensemble forecast schemes added (DSAEF_LTP-2), new similarity regions added (DSAEF_LTP-3), and both added (DSAEF_LTP- 4), respectively. Results showed that, after new similarity regions added into the model (DSAEF_LTP-3), the forecasting performance of the DSAEF_LTP model for heavy rainfall (accumulated precipitation ≥250 mm and ≥100 mm) improved, and the sum of the threat score (TS250 + TS100) increased by 4.44%. Although the forecasting performance of DSAEF_LTP-2 was the same as that of DSAEF_LTP-1, the forecasting performance was significantly improved and better than that of DSAEF_LTP-3 when the new ensemble schemes and similarity regions were added simultaneously (DSAEF_LTP-4), with the TS increasing by 25.36%. Moreover, the forecasting performance of the four experiments was compared with four operational numerical weather prediction models, and the comparison indicated that the DSAEF_LTP model showed advantages in predicting heavy rainfall. Finally, some issues associated with the experimental results and future improvements of the DSAEF_LTP model were discussed.     

Western North Pacific Tropical Cyclone Database Created by the China Meteorological Administration

This paper describes the access to, and the content, characteristics, and potential applications of the tropical cyclone(TC) database that is maintained and actively developed by the China Meteorological Administration, with the aim of facilitating its use in scientific research and operational services. This database records data relating to all TCs that have passed through the western North Pacific(WNP) and South China Sea(SCS) since 1949. TC data collection has expanded over recent decades via continuous TC monitoring using remote sensing and specialized field detection techniques,allowing collation of a multi-source TC database for the WNP and SCS that covers a long period, with wide coverage and many observational elements. This database now comprises a wide variety of information related to TCs, such as historical or real-time locations(i.e., best track and landfall), intensity, dynamic and thermal structures, wind strengths, precipitation amounts, and frequency. This database will support ongoing research into the processes and patterns associated with TC climatic activity and TC forecasting.     

IMPROVEMENT OF NUMERICAL SIMULATION OF TYPHOON RANANIM (0414) BY USING DOPPLER RADAR DATA

Typhoon Rananim (0414) has been simulated by using the non-hydrostatic Advanced Regional Prediction System (ARPS) from Center of Analysis and Prediction of Storms (CAPS). The prediction of Rananim has generally been improved with ARPS using the new generation CINRAD Doppler radar data. Numerical experiments with or without using the radar data have shown that model initial fields with the assimilated radar radial velocity data in ARPS can change the wind field at the middle and high levels of the troposphere; fine characteristics of the tropical cyclone (TC) are introduced into the initial wind, the x component of wind speed south of the TC is increased and so is the y component west of it. They lead to improved forecasting of TC tracks for the time after landfall. The field of water vapor mixing ratio, temperature, cloud water mixing ratio and rainwater mixing ratio have also been improved by using radar reflectivity data. The model’s initial response to the introduction of hydrometeors has been increased. It is shown that horizontal model resolution has a significant impact on intensity forecasts, by greatly improving the forecasting of TC rainfall, and heavy rainstorm of the TC specially, as well as its distribution and variation with time.     

热带气旋垂直倾斜度估算及在强度分析中的应用

该文给出了热带气旋暖心垂直倾斜度参数F_TC-VGP(TC vertical gradient parameter) 的定义。使用卫星微波遥感的温度垂直分布以及热带气旋最佳路径数据等资料,对2009—2011年典型热带气旋的暖心垂直倾斜度进行计算,并研究其与气旋强度变化的关系。结果表明:暖心垂直倾斜度参数F_TC-VGP变化与气旋自身强度变化有较好的对应性,该参数的变化能够明显地指示气旋强度增强、减弱以及强度突变, 能够指示环境场变化 (冷空气侵入) 等;其与气旋强度变化的周期不同,在强度增强或减弱阶段,倾斜度参数是经过快速适应性调整才完成了对热带气旋变化过程的贡献。因此,F_TC-VGP参数的计算和研究结果实现了对于热带气旋内部三维结构特征的定量描述,可以准确把握热带气旋强度和变化趋势,并对热带气旋的监测、预报具有一定业务参考价值。     

Improving Multi-Model Ensemble Forecasts of Tropical Cyclone Intensity Using Bayesian Model Averaging

This paper proposes a method for multi-model ensemble forecasting based on Bayesian model averaging (BMA), aiming to improve the accuracy of tropical cyclone (TC) intensity forecasts, especially forecasts of minimum surface pressure at the cyclone center (P_min). The multi-model ensemble comprises three operational forecast models: the Global Forecast System (GFS) of NCEP, the Hurricane Weather Research and Forecasting (HWRF) models of NCEP, and the Integrated Forecasting System (IFS) of ECMWF. The mean of a predictive distribution is taken as the BMA forecast. In this investigation, bias correction of the minimum surface pressure was applied at each forecast lead time, and the distribution (or probability density function, PDF) of P_min was used and transformed. Based on summer season forecasts for three years, we found that the intensity errors in TC forecast from the three models varied significantly. The HWRF had a much smaller intensity error for short lead-time forecasts. To demonstrate the proposed methodology, cross validation was implemented to ensure more efficient use of the sample data and more reliable testing. Comparative analysis shows that BMA for this three-model ensemble, after bias correction and distribution transformation, provided more accurate forecasts than did the best of the ensemble members (HWRF), with a 5%–7% decrease in root-mean-square error on average. BMA also outperformed the multi-model ensemble, and it produced “predictive variance” that represented the forecast uncertainty of the member models. In a word, the BMA method used in the multi-model ensemble forecasting was successful in TC intensity forecasts, and it has the potential to be applied to routine operational forecasting.     

国内外热带气旋强度变化研究现状

影响热带气旋强度变化的因子大致可以分为3类:环境气流与热带气旋环流的相互作用,下垫面与热带气旋环流的相互作用,热带气旋本身的内部结构变化。通过查阅近年来国内外相关文献,对热带气旋强度变化的研究进展进行了论述,为热带气旋强度研究提供参考。