单多普勒雷达反演涡旋风场方法分析

由于单雷达在探测台风临近、登陆时仍具有双部或多部雷达难以具备的优势,目前利用单多普勒雷达反演涡旋风场的方法并不多见,本文对能够反演涡旋风场的3种方法NIVAP(自然坐标系的积分VAP方法)、EVAPTC(反演热带气旋的EVAP方法)、GBVTD(地基速度追踪法)进行了比较,分别从每种方法的假设条件和数学方法以及结果精度来比较3种方法对涡旋风场的适用性。利用中尺度气旋Rankine模式模拟了纯涡旋性气旋、只有环境风场时、环境风场和辐合(辐散)同时存在时的径向速度场及风场,比较了各种情况下的反演风场和模拟风场的相似系数。在对模拟风场比较之后,选取了2014年"威马逊"台风登陆前1h、登陆时及登陆后1h的3个实测雷达资料,对比分析了3种方法反演实测资料的风场特征,3种方法中EVAPTC方法最好,GBVTD方法在应用中有一定的限制,NIVAP方法较差。     

江西省山地风场风速订正方法研究

利用气象站、探空及NASA再分析资料,对江西省4县山地风场的12座测风塔风速进行订正研究。研究结果表明:测风塔与气象站风速数据相关性较低,相关系数一般远小于0.45;测风塔与探空资料的风速相关系数可达到0.6以上,最高可达到0.8;NASA再分析资料可以作为江西山地风场风速订正参证数据,其与测风塔风速数据相关性较高,相关系数可达到0.54～0.77,大多数测风塔相关系数可达0.7左右。海拔高度小于1000 m的测风塔与NASA 50 m风速的相关系数明显高于其与NASA 850 hPa风速的相关系数,高度为1000—1200 m的测风塔与NASA 50 m风速和与NASA 850 hPa风速的相关系数相差不明显,高度大于1200 m的测风塔与NASA 850 hPa风速的相关系数明显大于其与NASA 50 m风速的相关系数。比值法订正效果略好于线性回归法的,订正后的风功率密度总体偏大。     

THE TEMPORAL AND SPATIAL DISTRIBUTION OF SEA SURFACE WIND ALONG THE COAST OF GUANGDONG PROVINCE IN CHINA

Temporal and spatial distribution characteristics of sea surface wind in Guangdong''s coastal areas were analyzed with data from four offshore observational stations between 2012 and 2015. The results are shown as follows: (1) The probability distribution of wind speed was basically consistent with Gaussian distribution characteristics; winds of Beaufort force 6 or higher were observed mainly in far offshore stations from October to March. (2) The probability distribution of wind direction was represented well by Weibull distribution. The deviation of wind direction of far station was relatively small for it was mainly controlled by monsoon over the South China Sea, while the near offshore station had a relatively large diurnal variation because of the influence of local synoptic systems such as sea-land breeze. (3) There were significant seasonal differences in wind speed and direction observed by different offshore observational stations. In strong wind seasons, the deviation of wind direction was relatively small while the deviation of wind speed was relatively large, and vice versa. In contrast with Class I station, the other three stations exhibited approximately normal distribution of wind direction and wind speed deviations. (4) Wind direction diurnal variation was moderate in windy periods, while it was obvious in relatively lower speed conditions. The deviation of wind speed in windy periods was generally greater because it was influenced by mesoscale weather systems for 10-20 h, and the influence was complicated, resulting in greater local differences in wind speed.     

INTERDECADAL VARIABILITY OF QUASI-BIENNIAL OSCILLATION OF STRATOSPHERIC ZONAL WIND

This study investigates the interdecadal variability of Quasi-biennial Oscillation (QBO) based on the sounding data in the stratosphere, ERA-40 and ERA-interim reanalysis data in the past 62 years. The QBO periodicity experiences a significant interdecadal variability; the longer (shorter) the mean period, the smaller (larger) the amplitude of variation is. The QBO amplitude varies in a cycle around 10 to 15 years and in an out-of-phase correlation with the period. In addition, there is an increasing trend of the QBO amplitude in 30 to 10 hPa, while a little declining trend in 70 to 40 hPa. The deviation of the QBO zonal wind extremum centers from the equator also shows interdecadal variability. The deviation location of the easterly core is generally in the reverse side to the westerly core, which means that when the easterly core is on one side of a hemisphere, the westerly core is on the other side.     

一次东北冷涡下槽后强风与飑线后向入流演变及成因分析

本文利用NCEP分析资料、多普勒雷达观测资料、常规气象观测资料以及数值模拟结果,对2016年7月30日发生在华北、辽宁附近的一次强飑线过程中后向入流的演变及成因进行研究。结果表明,此次飑线发生在中纬度新生冷涡槽前,低层有水汽辐合区和地面辐合线对应,且过程中伴有较强的对流有效位能释放。飑线后部中层(冷涡槽后)一直存在α中尺度西风大值带,此大风速带造成了上下层相反的水平涡度,并形成喇叭形环流结构,该结构不同于经典飑线结构。飑线后部水平方向上水平涡度分布不均匀,并形成水平涡度旋度上正下负的分布,即导致中层强风区上部上升运动、下部下沉运动,该下沉运动引发飑线中的后向入流和低层强风速带形成。在中层,飑线的后部边缘始终有较强的风速大值带伴随飑线的发展,该大值带的形成与对流强弱和非热成风涡度有关,对流过程中低层非热成风涡度为负,中上层非热成风涡度为正,导致飑线后部中层西风加速和低层西风减速,有利于后向入流的发展和飑线的维持,当对流减弱时,非热成风涡度与后向入流均减弱。文中给出了后向入流形成演变的概念模式。     

延庆-张家口地区复杂地形冬季山谷风特征分析

基于2016年12月—2017年2月和2017年12月—2018年2月两年冬季的近地面自动气象站逐时观测数据以及张家口探空数据分析延庆-张家口一带（包括张家口崇礼、赤城、海坨、小五台山区,延怀、怀涿、洋河、蔚县盆地以及北京延庆、昌平、怀柔部分平原地区）复杂地形的风场精细化时、空分布特征,揭示不同复杂地形下局地风场的时、空变化规律,加深对复杂地形动力、热力作用对近地面风场影响的认识,为冬季山区风场预报以及复杂地形数值模式改进提供参考。结果表明:晴朗小风天风持续性作为矢量平均风速和标量平均风速的比值,可以作为研究风场变化规律的重要参数。根据风持续性的日变化特征,可以将研究区域内所有站点分为10种类型,分别代表不同局地地形特征的影响,风持续与风向变化的相关也很强。研究区域主要有3种类型的地形风:斜坡风、峡谷风以及较大尺度的山区平原风。不同地形特征下的风场、风持续性存在明显不同的日变化特征,山风和谷风相互转化的时间也不同,山区最早,盆地次之,平原区最晚;山风时段持续时间较谷风时段长,风速小;晴朗小风天实测风反映了实际风场的特征,而排除环境背景风场,弱化地形动力作用后整个冬季的局地风作为理论山谷风,更能反映热力作用下的山谷风特征。      

Kinetic energy budget and moisture flux convergence analysis during interaction between two cyclonic systems: Case study

An analysis of the kinetic energy budget during a case of interaction between middle latitude and extratropical cyclones has been made in this work. Horizontal flux convergence constitutes a major energy sink. Generation of kinetic energy via cross-contour flow is a persistent source throughout the growth and decay periods. Dissipation of kinetic energy from subgrid to grid scales is an important source during the pre-storm period; it acts as a sink during the growth and decay periods. The major contribution to kinetic energy comes from a persistent upper tropospheric jet stream activity throughout the period of the cyclone development. The characteristics of moisture-flux components (divergent and rotational) along with precipitable water content for different tropospheric layers throughout the life cycle of our cyclone are also studied in this work. It is found that most of required humidity for our cyclone are initiated from Arabian Sea and then to some extent are reinforced over Gulf of Aden and east of central Africa and then by passing over Red sea enter to the south and south east of Mediterranean Sea. The rotational component of the moisture transport brings moisture from two regions; the first which is considered the main region is the Indian Ocean, Arabian Sea, Gulf of Aden and north east of Sudan. The second source region is the Atlantic and Mediterranean Sea. In the middle troposphere, the primary moisture source is found over central Africa, which in turn is traced to the Atlantic Ocean, the Indian Ocean, and the Arabian Sea. The upper-level moisture fluxes are weak and play a minor role over the area of interaction between two cyclones.     

北京山区与平原冬季近地面风的精细观测特征

利用2009—2018年冬季北京地区200多个自动气象站逐时10 m风速、风向观测数据,分典型区域（山区、山区与平原过渡区、平原区、城区）研究北京地区冬季近地面风的精细特征,并使用有完整记录的2 a（2017和2018年）冬季延庆高山区不同海拔高度10 m风逐时观测数据,多视角分析高山区不同海拔高度近地面风的特征和成因,以深刻认识北京地区复杂地形条件下冬季近地面风的特征和规律。结果表明:（1）北京地区冬季近地面平均风受西部北部地形、城市下垫面粗糙度和冷空气活动共同影响,平均风速沿地形梯度分布,山区高平原低,平原中又以城区风速最小;盛行西北风和北风,在城区东、西两侧盛行风出现扰流,在山区和过渡区一些地方还存在与局地地形环境明显关联的其他盛行风向。（2）4个典型区域冬季近地面风速日变化均表现为白天风速大于夜间,午间风速最大的“峰强谷平”单峰特征,这一特征的稳定性在城区高、山区低。（3）4个区域冬季弱风（< 1 m/s）频率为31%—42%,城区较高、山区较低;强风（> 10.8 m/s）频次则是山区多、城区少,强风风向主要表现为偏西—偏北,与冷空气活动密切关联;城区、平原区和过渡区偏南风频率均为极小,暗示北京“山区—平原”风模态在冬季是“隐式”的、不易被直接观测到。（4）近地面风的水平尺度代表范围在延庆高山区高海拔处明显大于低海拔处,海拔1500 m附近（平均的边界层顶高度）是延庆高山近地面风速日变化特征的“分水岭”,低于该海拔高度时近地面风速日变化表现为前述“峰强谷平”单峰特征,而高于该海拔高度时近地面风速日变化则呈现相反特征,即夜间大白天小、午间最小的“峰平谷深”特征,这是由边界层湍流活动的日变化及伴随的低层自由大气动量向边界层内下传所致。（5）延庆高山近地面风速大体上随观测高度而增大,高海拔站点日平均风速数倍于低海拔站点。白天—前半夜,海拔约2000 m的站点冬季盛行偏西风,风向变化不大,但风速为2—12 m/s;1000 m左右的低海拔站则风速比较稳定（< 6 m/s）,风向从午间至傍晚相对多变。      

京津冀地区低层局地大气环流的气候特征研究

本文采用2007-2016年的中尺度数值模拟结果和15个地面气象站观测资料,定义了局地风场表征风速,研究京津冀平原地区局地大气环流日变化的气候特征,并对区域大气污染及其输送的影响进行分析。此外,对2016年12月30日至2017年1月7日北京地区跨年大气重污染过程进行了个例分析。得到结论:京津冀平原地区低空风场变化是天气系统与局地大气环流共同作用的结果,山谷风环流致使太行山和燕山沿线平原地区大气边界层内的长年主导风向为偏北和偏南;太行山和燕山沿线地区山谷风环流本身呈顺时针旋转的日变化特征,夜间至早晨谷风转向山风,午后至夜间山风转向谷风;在午后谷风向山风转向期间,容易形成沿太行山东麓和燕山南麓、自南向东北的"弓形"气流输送通道,此气流输送通道在1月于21时左右形成,持续时间大约3 h,在7月于18时左右形成,持续时间可达9 h;冬季午后至晚间盛行谷风时,受山体的阻挡,污染物容易在山前累积,导致污染浓度增高;夏季同样的情况会发生在后半夜。     

基于小波长短期记忆网络的风电功率超短期概率预测

随着大规模的风电并网,风电所具有的间歇性与随机性对电力系统的稳定性产生了很大的影响,风电功率预测成为当前解决该问题重要的方式之一.本文利用长短期记忆（LSTM）网络良好的时序记忆特性,将小波分解技术与LSTM深度网络结合,提出基于小波长短期记忆网络的风电功率超短期概率预测模型.首先通过小波分解技术将原始时间序列进行平稳化处理,再建立各子序列样本的LSTM网络预测模型,借助最大似然估计法估计预测误差的高斯分布函数,最终实现对未来4 h时刻的风电功率概率区间预测.最后,采用中国东北某风电场数据对所提方法进行算例分析,结果表明,将小波分解与深度学习方法结合可以较好地提高预测的精度,提高概率预测的区间可靠性.