结合公众不同出行方式和广州本地气候,分析影响公众出行的主要气象要素及各要素对公众公交地铁、骑行、步行、驾车4种出行方式的影响.开展公众不同出行方式所关注的气象要素调查并收集到4 935份问卷,根据问卷结果和分析情况设计出了不同出行方式的出行气象提醒规则.在全行程气象影响分析的基础上,将提醒规则分为3个层次,天气现象提醒-影响提醒-防御提醒,层层递进,为用户展示全行程天气情况.

公众气象服务是政府公共服务的重要组成部分,而微博作为新时代下的新媒体服务,在广州市气象局的日常气象信息服务中发挥了重要作用.通过对微博渠道积累的数据及用户需求进行文本数据分析挖掘,发现许多影响微博受欢迎程度的用户偏好特征,运用联合分析得到各个特征的重要性排序,并以此提出新媒体运维指引以及提升公共气象服务的合理建议,从而在扩大微博影响力的同时增强广州市气象局公共气象服务的品牌影响力.

根据2016年广州天气微博矩阵运维数据探讨微博圈话题聚合效应,结果表明:(1)通过气象微博矩阵进行气象服务,可以促使矩阵成员之间加强联动,扩大微博运营效果;(2)话题聚合是广州天气微博矩阵运营的一个重要手段,话题聚合可以整合服务资源,增加矩阵各成员的曝光度,带动整体影响力的增长;(3)微博矩阵的建设可以直接拉动博文阅读量、粉丝量的增长,增强天气官博的互动力,从而提升矩阵成员的整体影响力.

天气预警智能推送系统的建设是以微信模板消息功能特点为基础,同时结合天气预警推送需要,实现了基于位置的天气预警信息点对点发布.根据微信提供的模板消息,该系统可针对预警信息、突发应急信息、强灾害性天气影响等采用不同的模板来实现信息的智能推送.系统设计后台采用多线程处理技术,提高预警推送效率,确保预警信息在10 min之内到达用户端,并且利用用户ID绑定等方式实现用户对预警信息的分区自主定制.基于微信模板消息的天气预警智能推送系统是精细化预警发布方式的重要探索,很好地提高天气预警信息传播覆盖面和传播效率.

从微博互动的角度对新政务服务形势下的气象微博运维策略动作的分析得出:(1)以天气预提醒和过程追踪的方式提供微博气象服务,以话题聚焦热点天气问题,适度结合网络语言进行产品包装,形象化描述天气现象、贴合生活出行进行温馨提醒.(2)利用大数据统计分析,针对公众日常活动和活跃时间自动推送相关度高的预报信息,实时检索、智能推送图文并茂的灾害性天气提醒、预警信号等时效性的信息.(3)利用评论和私信回复有效提升气象微博互动,评论回复以官博互动为主,小编互动为辅,私信回复利用粉丝服务平台接入气象数据库,提供自定义回复与官博私信回复.

从微博影响力评价指标的3个方面:传播力、服务力和互动力出发对广州天气微博运维情况进行分析,结果表明:(1)广州天气采用主动推送与人工采编相结合的方式运维.实用性及时效性强的信息采用自动发布的方式,充分发挥预警类信息的防灾减灾作用;科普类及天气预报类信息采用人工采编的方式,恰当运用网络语言,做公众喜爱的气象服务产品.(2)实用性强的信息固定时间发布培养公众定时获取气象信息的使用习惯;而具备一定趣味性的信息如科普类信息,抓住用户的碎片化阅读时间发布,在天气转折用户关注的情况下及时进行实况追踪、实况播报.(3)互动力的提升包括官博与粉丝的互动及小编帐号的互动,联合政府部门扩大气象官博的影响力,官博及认证小编对粉丝的互动可以及时满足粉丝的个性化需求,特别小编账号让互动变得灵活及主动,粉丝也非常愿意回应.

利用微信的“点对点”、“互动交流”的优势,探索微信在开拓气象服务发布新渠道的发展思路,提升公共气象服务能力.广州天气微信挖掘微信平台本质,按“有用”且“能玩”的建设思路,以用户感知习惯为导向,以格点预报、自动站数据、精细化雷达等丰富的产品资源为基础,打破微信传统布局,通过前台菜单与后台管理的巧妙配合,实现“点菜”功能,成功创建全国首个交互式气象服务微门户,并结合用户需求对气象数据进行挖掘和包装,创新设计出“我的晴雨钟”、“实况大PK”等好用又好玩的产品,实现对用户的个性化定制服务,将气象服务从“有什么给提供什么”向“用户需要什么就给用户提供什么”的转变,满足用户的个性化需求与娱乐需求.

Precipitation data of Guangzhou with the historical synoptic maps were studied,to investigate the climate features and the weather patterns for torrential rainfall(hereafter as TR) events in Guangzhou during the first flood season(April to June,hereafter as FFS) in recent 59 years.Results show that 1) the accumulated rainfall of TR-FFS has the same increasing trend with the total accumulated rainfall in FFS and the contribution of the TR-FFS shows a wave pattern;2) the proportion of the accumulated TR during May and June in the total accumulated rainfall of FFS,has a 10-year periodic variation,indicating high probability of TR in May in the next 10 years;3) the frequency of TR-FFS contributes 52.1% to that of the whole year TR with 4-8 year oscillation in wavelet power spectrum;4) TR-FFS occur when cold and warm air mass converge in Guangzhou with even intensity,and single air mass dominating south China usually leads to less TR-FFS in Guangzhou.

利用常规气象观测资料、NCEP1°×1°再分析格点资料以及多普勒雷达资料,对"狮子山"持续性强降水进行分析,找出了其相应的天气环流特征.

Based on NCEP reanalysis and conventional observational data,Analyses are performed on the features and causes of a province - wide low - temperature and rainy weather during 2010 Spring Festival in Guangdong.Results show that the weather has characteristics of long-lasting process,wide range of influence and great impact with abnormally low daily-average temperature and frequent precipitation;stable maintenance of abnormally stronger Ural blocking high and Mongolia transversal trough,anomalously weaker East - Asia trough and stronger subtropical high as well as the smooth atmospheric circulation over low-and-mid-latitude areas contribute to the favorable atmospheric circulation.Frequent invasions of shallow or low-level cold air mass and long - time lingering of wind shear over middle and northern Guangdong directly lead to such cold and rainy weather.

Diagnostic analysis on a non-typical cold air associated torrential rain in Southern Guangdong is performed by using conventional observational data,automatic weather station data and the NCEP reanalysis data(1×1 deg) as well as the Doppler radar data.Results show that(1) obvious mesoscale characteristics can be identified in this case;(2) vertical upward motion began in advance of the occurrence of the heavy rain,while the low level moisture convergence agreed well in time with the intensity of precipitation;(3) accumulation of the instability energy beforehand is necessary,the invasion of the cold front is the trigger mechanism,and the distribution of the convective available potential energy(CAPE) in a degree indicates the probable torrential rainfall areas;(4) the intensity and the development of radial velocity convergence of radar can give us some indications of the evolution of the convective echoes.

Strong cold air is one of the high impact weathers in Guangzhou in November.This paper analyzes the observational data of daily mean temperature,maximum temperature,minimum temperature and daily rainfall amount from 5 stations of Guangzhou in November during 1961 to 2009,together with the NCEP reanalysis data(2×2 deg).Results show that no cold wave but 8 strong cold air events are found during this period according to the cold wave criteria of Guangdong.Among them,the one arriving at Guangzhou on November 27,1987 is the strongest.In this case,cold air outburst southward steered by meridionally turning of transversal trough at 500 hPa,with the tropical cyclone NINA approaching coast and sustaining for 24 hours,which supply favorable moisture and convergence for the precipitation.Significant gradient wind between tropical systems and the cold air mass accelerated the southward cold air.Development of the low trough in western China causing a warming up in Guangzhou before the cold air outbreak,also cause a relatively significant drop in temperature.

利用差值分析方法对广州国家基本站新旧址的常规同步观测气象要素资料进行分析,结果表明:(1)萝岗新站平均气温、最高气温、最低气温均小于五山旧站,年平均气温差值为0.9℃,年平均最高气温差值为0.4℃,年平均最低气温差值为1.1℃.(2)萝岗新站的平均风速明显大于五山旧站,萝岗新站全年以偏北风为主,其中春夏季节的偏南风也占一定的比例,风向的季节变化不明显.(3)高温天气过程的最高气温差异不明显,风速较大时,最大差异为0.6℃.冷空气过程中,平流降温占主导地位时,新旧测站的最低气温差异约0.7℃；平均最低气温差异随晴空辐射加强而增大,平均差异达-2.7℃,极端差异达4.2℃.(4)萝岗新址2010年观测到极端最低气温-1.2℃,破了广州市极端最低气温0℃的记录.

利用NCEP再分析资料和常规气象观测数据以及大气成分观测数据,对2009年11月24~29日广州市一次持续灰霾天气过程做了初步的分析。结果表明,此次灰霾天气过程在850 hPa为反气旋环流控制,灰霾的严重程度与850 hPa风速的大小密切相关;地面形势则为弱高压脊或脊后回流型,气压场多为均压场,海平面气压值通常小于1 017 hPa,灰霾的严重程度与弱压脊的位置及海平面气压大小密切相关;按气象标准当达到了灰霾日标准后,能见度可作为衡量灰霾严重程度的指标;此次灰霾天气过程中,SO2污染物在影响能见度中起到了主导作用。