11个S2S模式对MJO预报效果的评估分析

利用次季节—季节预报研究计划(Subseasonal to Seasonal Prediction Project, S2S)的多模式产品集,系统评估了产品集中11个模式对MJO的实际预报技巧。如果以距平相关系数ACC为0.5作为有效预报技巧的阈值,S2S各模式的MJO实际预报时效为8～32 d。S2S各模式预报普遍低估了MJO的振幅强度,且预报的MJO传播速度偏慢。通过分析发现,在一个集合预报系统中,集合离散度与均方根误差越接近,它的MJO预报技巧越高。此外,分析S2S各模式MJO预报技巧对起报时间、季节和起报时MJO信号强弱的敏感性发现,当起报时间为冬季且起报时MJO为强信号时,MJO的实际预报技巧较高。     

次季节—季节(S2S)预测数据分析与应用

S2S（Sub seasonal to Seasonal）国际合作计划的主要目标是提升次季节〖CD*2〗季节多模式预测能力，重点关注0～60 d的预报预测技巧。中国气象局承担了S2S数据的归档备份任务，收集参与该计划的11个中心S2S模式数据，本文梳理了该数据资源的基本情况，介绍了数据的模式、要素及文件，研究了数据的读取和分析方法，评估了数据的正确性、完整性和时效性，运用模式回算数据，计算候、周、旬时间尺度的预测值和距平值，开展了相关数据产品的可视化工作，对照同期ECMWF距平图形产品进行分析，推动S2S数据产品在中国气象局预报预测业务中应用。     

“集合统计量集成法”在湖北短期暴雨预报中的应用研究

利用2013年11月—2014年11月ECMWF全球集合预报的51个成员降水预报资料和湖北省78个国家气象站逐日降水实况,对集合平均值、分位值、概率匹配平均值、众数(Mode值)等10种集合统计量在湖北省不同预报分区内的降水预报效果分别进行检验评估。在此基础上,采用在不同降水量级上选取TS评分最优的集合统计量的原则,设计出适用湖北降水预报的最佳集合统计量集成方案,并检验了该方法在2015年和2016年6—8月湖北降水预报的应用效果。结果表明:将集合统计量集成法应用于湖北降水预报时,集合统计量集成方案应随着预报分区的改变而改变;改进后的ECMWF集合统计量集成方案对湖北72 h内大雨及以上降水预报的TS评分均有不同程度的提高,且空报率和漏报率总体上有所降低;与ECMWF确定性预报相比,ECMWF集合统计量集成预报产品对湖北24 h内各降水量级的预报均优于ECMWF确定性预报,且对湖北72 h内的暴雨预报准确率均高于ECMWF确定性预报。     

两个集合预报系统对秦岭及周边降水预报性能对比

利用欧洲中期天气预报中心 (ECMWF)、美国大气环境预报中心 (NCEP) 集合预报系统 (EPS) 降水量预报资料,CMORPH (NOAA Climate Prediction Center Morphing Method) 卫星与全国3万个自动气象站降水量融合资料,基于技巧评分、ROC (relative operating characteristic) 分析等方法,对比两个集合预报系统对秦岭及周边地区的降水预报性能。结果表明:两个系统均能较好表现降水量的空间形态,对于不同量级降水,ECMWF集合预报系统0~240 h控制及扰动预报优于NCEP集合预报系统,但NCEP集合预报系统264~360 h预报时效整体表现更好; ECMWF集合预报系统0~120 h大雨集合平均优于NCEP集合预报系统,两个系统集合平均的预报技巧整体低于其控制及扰动成员预报,这种现象ECMWF集合预报系统表现更为显著; ECMWF集合预报系统降水预报概率优于NCEP集合预报系统。ROC分析显示,随着预报概率的增大,ECMWF集合预报系统在命中率略微下降的情况下,显著减小了空报率,NCEP集合预报系统则表现出高空报、高命中率。     

我国近海洋面10 m风速集合预报客观订正方

利用2013—2015年ECMWF集合预报10 m风场及我国沿岸和近海88个代表站点风速实况观测,建立基于ECMWF集合预报众数的我国近海洋面10 m风速客观订正方法。集合预报众数正确率及稳定性高于中值及平均值,因此基于集合预报众数,综合考虑历史数据的预报概率及集合预报各个成员的分布情况进行客观订正,可以提高订正效果。订正后的6~7级、8~9级风速偏小的误差及TS评分有明显改进,其中72~120 h预报时效的8~9级风速预报的TS评分由0.04增加到0.44,能够有效提高中长期时效大量级风速的预报能力。订正的风速产品对于我国近海冷空气及台风大风天气过程有较好的预报效果。     

基于TIGGE资料的集合成员优选方法

利用2007年6月8日—8月31日东亚地区TIGGE集合预报资料中欧洲中期天气预报中心(European Centre for Medium-range Weather Forecasts,ECMWF)和英国气象局(United Kingdom M et Office,UKM O)两个中心的地面2 m气温资料进行集合成员优选研究。结果表明,对于24~96 h预报,集合成员优选方法能够较好地选出预报技巧较高和预报技巧较低的集合成员。个例分析表明,在极端天气出现的地区,优选集合平均的预报优势较为明显。对比ECMWF和UKMO的集合成员优选结果发现,ECMWF的预报效果优于UKMO的预报效果。     

我国近海洋面10 m风速集合预报客观订正方法

利用2013—2015年ECMWF集合预报10 m风场及我国沿岸和近海88个代表站点风速实况观测,建立基于ECMWF集合预报众数的我国近海洋面10 m风速客观订正方法。集合预报众数正确率及稳定性高于中值及平均值,因此基于集合预报众数,综合考虑历史数据的预报概率及集合预报各个成员的分布情况进行客观订正,可以提高订正效果。订正后的6~7级、8~9级风速偏小的误差及TS评分有明显改进,其中72~120 h预报时效的8~9级风速预报的TS评分由0.04增加到0.44,能够有效提高中长期时效大量级风速的预报能力。订正的风速产品对于我国近海冷空气及台风大风天气过程有较好的预报效果。     

时空投影模型(STPM)的次季节至季节(S2S)预测应用进展

随着数值天气预报技术和季节动力预报系统的发展,短期天气预报及长期气候预测的能力持续提高,然而介于两者之间的次季节至季节(S2S,两周至三个月)预测技巧偏低,成为当今气象学界和业务服务的难题。南京信息工程大学国家特聘专家李天明教授团队于2012年研发了基于时空投影技术的统计预报模型(STPM),成功地对中国大陆降水和气温距平,以及区域极端降水、夏季高温、冬季低温和西太平洋台风群发事件等高影响天气进行提前10~30 d的预报,并在国家气候中心及多个省份开展了业务应用。STPM也成功应用于台湾春雨预报、南海季风爆发和ENSO预测等季节至年际变化的预测。本文对S2S预测的理论基础、STPM的发展和应用进行了完整的介绍,并讨论了S2S预测业务中所面临的挑战和未来展望。     

基于TIGGE资料的流域概率性降水预报评估

利用2008年7月1日至8月6日TIGGE-CMA资料存储中心的ECMWF、NCEP和CMA等业务中心1～10天的集合预报降水结果,结合淮河流域上游大坡岭—王家坝流域内19个站点的降水观测资料,对流域内的日降水预报效果进行了基于降水等级划分的确定性TS评分、概率性Brier评分以及考虑所有降水强度概率的百分位降水评估,并对2008年7月22—23日的强降水过程的预报效果进行了重点评估分析,探索了多模式概率预报降水面向流域的评估方法。结果表明,超级集合的TS评分和Brier评分优于单个中心的集合预报平均,集合平均由于平滑作用削弱了对长预报时效较强降水的预报能力;三套集合预报都体现部分成员具有捕捉实际降水的多种可能性;流域面雨量和单站百分位的分析表明:随着预报时效的延长,强降水的预报能力逐渐减弱,而超级集合由于考虑了更多的降水可能性,预报强降水的量级和空间分布同观测更为接近。     

面向台风暴雨的集合预报成员优选订正技术研究及应用试验

本文以2013—2015年主要登陆台风暴雨过程为研究对象,利用ECMWF降水和台风路径集合预报以及中央气象台实时业务台风中心定位资料,在统计分析的基础上,提出一种业务上可用的针对单模式集合预报的台风降水实时订正技术(简称集合成员优选技术)。结果表明,在登陆台风暴雨过程预报中,集合成员优选技术对改进集合统计量降水产品有明显的帮助,并较ECMWF确定性预报产品有一定优势;该方法对改进短期时效预报产品的效果优于中期时效预报,对大暴雨评分的改进高于暴雨和大雨评分。另外,本文基于概率匹配平均(Probability Matching average,PM)和融合(FUSE)产品的计算原理,提出融合匹配平均(Fuse Matching average,FM)产品,结果表明,对36 h时效预报,优选10~15个成员的PM产品TS(Threat Scores)评分可达最优,大暴雨评分较确定性预报提高近10%;对60和84 h时效预报,FM产品大暴雨评分较确定性预报提高超过20%。     

Evaluation of the ECMWF 32-day ensemble predictions during 2009 season of western North Pacific tropical cyclone events on intraseasonal timescales

The performance of the ECMWF 32-day ensemble predictions of western North Pacific tropical cyclone events (formation plus track) made once a week during the 2009 season is evaluated with the same procedures as for the 2008 season. Seventeen of the 23 tropical cyclones during the 2009 season occurred during multiple storm scenarios that are more difficult to predict, and many of the deficient track predictions involved unusual and rapidly changing tracks that typically involve interactions with adjacent synoptic circulations that are not predictable on intraseasonal timescales (10–30 days). Such incorrect predictions of the duration and tracks of these multiple cyclones were found to degrade the performance in predicting subsequent tropical cyclone formations and tracks during the 32-day integration. Predominantly northward tracks throughout the life cycle tended to be less predictable on intraseasonal timescales. Given these caveats, the overall performance of the ECMWF ensemble for the 12 typhoons was more successful than during the 2008 season. However, the performance for three tropical storms during the 2009 season was less successful due to the difficult track forecast scenarios. A surprisingly good performance was found in predicting the formation location and early track segments of eight minimal tropical storms or tropical depressions. The less satisfactory aspect for many of the late season tropical depressions was that the ECMWF ensemble continued to predict member vortices for extended periods after the system had actually dissipated.     

Predictability of tropical cyclone events on intraseasonal timescales with the ECMWF monthly forecast model

The objective of this study is to provide evidence of predictability on intraseasonal time scales (10–30 days) for western North Pacific tropical cyclone formation and subsequent tracks using the 51-member ECMWF 32-day forecasts made once a week from 5 June through 25 December 2008. Ensemble storms are defined by grouping ensemble member vortices whose positions are within a specified separation distance that is equal to 180 n mi at the initial forecast time t and increases linearly to 420 n mi at Day 14 and then is constant. The 12-h track segments are calculated with a Weighted-Mean Vector Motion technique in which the weighting factor is inversely proportional to the distance from the endpoint of the previous 12-h motion vector. Seventy-six percent of the ensemble storms had five or fewer member vortices. On average, the ensemble storms begin 2.5 days before the first entry of the Joint Typhoon Warning Center (JTWC) best-track file, tend to translate too slowly in the deep tropics, and persist for longer periods over land. A strict objective matching technique with the JTWC storms is combined with a second subjective procedure that is then applied to identify nearby ensemble storms that would indicate a greater likelihood of a tropical cyclone developing in that region with that track orientation. The ensemble storms identified in the ECMWF 32-day forecasts provided guidance on intraseasonal timescales of the formations and tracks of the three strongest typhoons and two other typhoons, but not for two early season typhoons and the late season Dolphin. Four strong tropical storms were predicted consistently over Week-1 through Week-4, as was one weak tropical storm. Two other weak tropical storms, three tropical cyclones that developed from precursor baroclinic systems, and three other tropical depressions were not predicted on intraseasonal timescales. At least for the strongest tropical cyclones during the peak season, the ECMWF 32-day ensemble provides guidance of formation and tracks on 10–30 day timescales.     

OPERATIONAL ENSEMBLE FORECASTING AND ANALYSIS OF TROPICAL CYCLONES OVER THE WESTERN NORTH PACIFIC (INCLUDING THE SOUTH CHINA SEA)

Based on the tropical cyclone data from the Central Meteorological Observatory of China, Japan Meteorological Agency, Joint Typhoon Warning Center and European Centre for Medium-Range Weather Forecasts （ECMWF） during the period of 2004 to 2009, three consensus methods are used in tropical cyclone （TC） track forecasts. Operational consensus results show that the objective forecasts of ECMWF help to improve consensus skill by 2%, 3%-5% and 3%-5%, decrease track bias by 2.5 kin, 6-9 km and 10-12 km for the 24 h, 48 h and 72 h forecasts respectively over the years of 2007 to 2009. Analysis also indicates that consensus forecasts hold positive skills relative to each member. The multivariate regression composite is a method that shows relatively low skill, while the methods of arithmetic averaging and composite （in which the weighting coefficient is the reciprocal square of mean error of members） have almost comparable skills among members. Consensus forecast for a lead time of 96 h has negative skill relative to the ECMWF objective forecast.     

Yearly evolution of Euro-Atlantic weather regimes and of their sub-seasonal predictability

It is often assumed that weather regimes adequately characterize atmospheric circulation variability. However, regime classifications spanning many months and with a low number of regimes may not satisfy this assumption. The first aim of this study is to test such hypothesis for the Euro-Atlantic region. The second one is to extend the assessment of sub-seasonal forecast skill in predicting the frequencies of occurrence of the regimes beyond the winter season. Two regime classifications of four regimes each were obtained from sea level pressure anomalies clustered from October to March and from April to September respectively. Their spatial patterns were compared with those representing the annual cycle. Results highlight that the two regime classifications are able to reproduce most part of the patterns of the annual cycle, except during the transition weeks between the two periods, when patterns of the annual cycle resembling Atlantic Low regime are not also observed in any of the two classifications. Forecast skill of Atlantic Low was found to be similar to that of NAO+, the regime replacing Atlantic Low in the two classifications. Thus, although clustering yearly circulation data in two periods of 6 months each introduces a few deviations from the annual cycle of the regime patterns, it does not negatively affect sub-seasonal forecast skill. Beyond the winter season and the first ten forecast days, sub-seasonal forecasts of ECMWF are still able to achieve weekly frequency correlations of r = 0.5 for some regimes and start dates, including summer ones. ECMWF forecasts beat climatological forecasts in case of long-lasting regime events, and when measured by the fair continuous ranked probability skill score, but not when measured by the Brier skill score. Thus, more efforts have to be done yet in order to achieve minimum skill necessary to develop forecast products based on weather regimes outside winter season.

     

辽宁省春季透雨集合预报误差分析

利用2016—2018年4月1日至6月30日三个全球数值预报业务中心(CMA、ECMWF和NCEP)的24 h降水集合预报资料和辽宁省降水观测资料,采用TS评分、预报偏差B、Talagrand分布以及BS评分等方法对辽宁省春季透雨(4—6月)CMA、ECMWF和NCEP三套全球集合预报结果进行对比分析.结果表明:三个集...     

湖南6月区域持续性暴雨概率预报模型及应用

利用1979—2016年6月EAR5再分析资料,选取湿热力平流参数、热力螺旋度、散度通量、水汽散度通量和热力波作用密度5个综合因子,采用核密度估计方法,基于TS评分最优为检验标准筛选确立最优因子和权重组合,构建了湖南区域持续性暴雨概率预报模型,并进行了独立样本检验与业务试用。结果表明:2017—2019年独立样本回代检验,平均TS评分达到29.9%,相比于欧洲中期天气预报中心(ECMWF)细网格(平均TS评分为22.4%)为正技巧。在2021年、2022年汛期两次区域持续性暴雨个例的预报试验中,提前24 h的暴雨预报优于ECMWF、CMA-GFS等大尺度模式和CMA-SH、CMA-GD等区域中尺度模式,对湖南区域持续性暴雨有较强的预报能力。     

东亚地区冬季地面气温延伸期概率预报研究

利用TIGGE资料中的ECMWF、NCEP、UKMO三个中心集合预报系统以及由此构成的多中心集合预报系统所提供的地面2 m气温10~15 d延伸期集合预报产品,建立贝叶斯模式平均(Bayesian Model Averaging,BMA)概率预报模型,对东亚地区冬季地面气温进行延伸期概率预报研究。采用距平相关系数、均方根误差、布莱尔评分、等级概率评分等指标分别对BMA确定性结果与概率预报进行评估。结果表明,BMA方法明显地改进了原始集合预报结果,预报技巧优于原始集合预报,且多中心BMA预报优于单中心BMA预报,最佳滑动训练期取35 d。BMA预报为气温的延伸期概率预报提供了更合理的概率分布,定量描述了预报的不确定性。     

2016-2020年广东省定量降水预报检验评估

对2016-2020年全球模式ECMWF和区域模式GZ_GRAPES、基于模式的解释应用和广东省气象局发布的定量降水预报（QPF）进行检验和评估。结果表明:ECMWF和GZ_GRAPES模式对一般性降水预报技巧在逐年提升,对大雨或以上的降水预报技巧的提升缓慢。GZ_GRAPES对大雨以上降水的预报技巧和定量降水预报的精细时空分布均优于ECMWF,区域模式更易预报出中小尺度降水信息。分类暴雨评定表明,模式对台风暴雨预报最好、锋面暴雨次之、季风暴雨预报最差。模式的暴雨预报落区偏小、低估明显,预报员通过经验订正明显提升了暴雨预报评分,其中季风暴雨的订正量最大,但存在预报范围偏大、空报较高的问题。基于ECMWF集合预报的解释应用与预报员的定量降水预报能力相当,降水越强,解释应用技术的优势越明显,但对季风暴雨也存在严重低估或漏报。目前降水精细时空分布、季风暴雨、极端性暴雨等依然靠预报员的经验订正为主,随着集合预报模式和区域高分辨率模式能力的提升,将预报经验客观化并与数值预报解释应用技术结合是提升QPF的一个方向。      

配料法在宁夏暴雨预报中的应用

选取2000—2015年发生在宁夏23次由深厚湿对流引起的区域性暴雨天气过程为研究对象,研究确定暴雨预报的配料方案,采用2016年6—8月宁夏25个常规站和947个自动站逐小时降水量、ECMWF 72 h内模式资料,从统计分析和个例剖析两方面探讨配料法暴雨预报效果,结果表明:(1)通过对比2016年6—8月ECMWF暴雨预报与配料法暴雨预报发现,不同预报时段,配料法暴雨预报均优于ECMWF暴雨预报;(2)配料法能准确预报所选两次暴雨个例降水的中心强度、落区及变化趋势,但强度和落区较实况偏强、偏大;(3)配料法对六盘山区暴雨过程弱降水预报效果较差,对贺兰山沿山暴雨过程存在降水空报。