利用FY-2C卫星数据反演云辐射特性

本文利用FY-2C静止卫星提供的可见光、中红外和热红外观测数据, 开展了水云光学厚度、粒子有效半径和云顶温度的云参数遥感探测理论和反演方法研究。基于FY-2C可见光、中红外(3.75 μm)与热红外(11 μm)通道辐射率对云光学厚度、 云滴有效半径、云顶温度辐射参数的敏感性分析, 提出三通道同时反演云的光学厚度、云滴有效半径及云顶温度的迭代方案; 通过个例分析进行了云参数反演试验, 并将结果与MODIS的云反演产品进行了对比, 最后对反演误差进行了分析。主要结论如下:(1) 个例反演得到的云参数与各通道探测数据有着较好的对应关系, 迭代计算标准偏差在允许的计算精度范围内(<0.89%), 反演结果具有合理性; (2) 通过与MODIS云反演产品的对比可以看到, 两者云光学厚度、云滴有效半径的均值和直方图分布都非常一致, 而MODIS的云顶温度比FY-2C反演值要高, 考虑到FY-2C的 11 μm通道测量的辐射值与MODIS相比偏小, 因此认为我们的反演方法与MODIS方法的精度是相当的。     

基于模板匹配的云底高度估计

本文提出一种将主动卫星遥感云底高数据扩展到被动遥感卫星视场的新方法。在分析云顶高度(CTH)和云水路径(CWP)对云垂直分布影响的基础上,提出用CTH和CWP两项参数为基的模板匹配法,结合加权最近邻插值估计云层厚度,并用成熟的CTH反演产品减去云层厚度,得到云底高度(CBH)。然后引入小波去噪的方法抑制对云底高度估计的局部高频涨落。与基于云类型的估计方法相比,该方法在20~300km范围内与其估计效果相当,且在170~240km范围内误差更小。考虑到基于云类型的估计方法在CloudSat卫星和MODIS云分类结果上存在差异,本文方法的适用性更好。     

利用美国NOAA卫星遥测云顶高度的数值试验

无论从气象角度还是气候角度讲,对卫星云辐射测值进行反演判释研究,从而提取诸如云顶高度、云顶温度、云的放射率、云厚以及云的含水量等参数都十分必要。本文主要介绍用NOAA卫星红外和微波通道测值,按其辐射率差比方法,定量判释两层云高的初步数值试验结果。经与地面实况对照分析表明:适当选择两组探测通道组合可以遥感两层云高,但还不能判释出中云和低云并存时各自的高度。     

FY 2F卫星和毫米波云雷达云高观测的个例对比分析

风云二号静止卫星上装载有可见光、水汽、中长波红外等探测通道,其中红外通道资料可提供卫星云顶温度数据。基于FY-2F静止卫星云顶温度资料,结合局地实时探空数据对北京南郊和朝阳站点上空云层进行云高反演,并展开与地基毫米波云雷达探测云顶高关系的对比,分析3种不同云厚(薄云、适中、厚云)条件下的云高观测结果。研究结果表明,二者云顶高匹配度受几何云厚的影响,其吻合度呈现出厚云最佳,薄云最差的特征。     

红外高光谱资料模拟大气廓线反演对云的敏感性

用不同云顶高度和不同有效云量时的星载红外高光谱观测值,模拟大气温湿垂直廓线反演对云参数的敏感性。用特征向量统计反演法按照云顶高度和有效云量分类反演了大气温湿廓线,统计了不同云高（200,300,500,700和850hPa）及云高有50hPa误差时温度、水汽混合比反演的均方根误差随有效云量的变化。结果表明,随着有效云量的增大,云顶以下各高度层上的温度、水汽混合比反演误差都明显增大。云高有50hPa误差较准确已知而言,温度和水汽混合比的反演误差增大,但温度反演对云高误差的敏感性比水汽反演要高。     

反演卫星资料在航空气象保障中的应用

讨论了用卫星资料反演资料Tbb和反照率计算云顶高度场的方法.并选取两个时次的卫星资料,同时计算昆明管制区的温度廓线,结合不同云的物理性质,得到云顶模拟高度场,最终制作了重要云的平面定位图和三维空间模拟图.     

反演卫星资料在航空气象保障中的应用

讨论了用卫星资料反演资料Tbb和反照率计算云顶高度场的方法。并选取两个时次的卫星资料，同时计算昆明管制区的温度廓线，结合不同云的物理性质，得到云顶模拟高度场，最终制作了重要云的平面定位图和三维空间模拟图。     

用FY-2C/D卫星等综合观测资料反演云物理特性产品及检验

云的宏微观物理特性参数无论对天气、气候还是人工影响天气的研究和业务都有十分重要的应用价值.基于FY-2C/D静止卫星遥感观测,融合高空和地面等其它观测资料,研发了近10种云宏微观物理特性参数的反演技术方法,并实现业务化运行.简单介绍反演得到的云顶高度、云顶温度、云过冷层厚度、云暖层厚度、云底高度、云体厚度、云光学厚度、云粒子有效半径和云液水路径等近10种云宏微观物理参数产品的物理意义、反演技术方法和业务流程等;对主要云参数产品,利用最新获得的Cioudsat云卫星实测结果进行了对比检验和可用性分析;将反演产品同MODIS反演的同类产品进行对比分析,发现两者具有较好的一致性.     

基于MODIS云宏微观特性的卫星云分类方法

利用MODIS云光学厚度、云粒子有效半径、云顶高度、云相态等产品,以及表征6种云类的云光学厚度、云粒子有效半径、云顶高度、云相态的特征值,采用最小距离分类法和多阈值判识法相结合,对卫星观测像元的云进行分类,包括层云、层积云、积云、积雨云、雨层云、高积云/高层云、卷云以及卷云伴随高积云或高层云的多层云、卷云伴随层云或层积云的多层云、高积云或高层云伴随层积云或层云的多层云10类。2008年、2013年卫星分类结果与地面站云类观测对比,达到60%的一致性;将相同时间的地面小时降水量与分类结果叠加显示,出现降水处多为雨层云或积雨云。     

利用卫星产品评估数值预报模式云层垂直结构

对云层垂直结构的准确参数化描述是数值天气预报模式准确计算长短波辐射通量、辐射加热率廓线、云反射率、云辐射效应等参数的重要基础,但地基观测数据无法对模式预报的云层垂直分层情况进行验证。文章基于卫星资料Collection5版本的MODIS云产品MOD06,利用国际上能够较准确判别云层垂直分层的一个新算法,以反演的高、中、低云发生频率和云顶气压结果,评估美国国家环境预报中心(NCEP)北美中尺度模式NAM的云层垂直结构。2006年7～10月北美地区(153°～48°W,12°～62°N)的评估结果表明:①卫星反演和模式预报的高、中、低云的云量区域分布比较相似,尤其是高云。热带太平洋地区模式预报高云量大于卫星反演值。模式预报的低云在墨西哥及北美大陆、大西洋地区更多。②卫星反演和模式预报中云发生频率的差异最小,模式预报高云和低云发生频率峰值比卫星反演的峰值更大,且云顶出现的高度更高。③模式预报中云量和低云量的纬度平均值比卫星反演的高,尤其是低云量。NAM的云参数化有待于进一步改进。     

基于2014～～2017年Ka毫米波雷达数据分析北京地区云宏观分布特征

本文利用2014年1月至2017年12月Ka毫米波雷达数据对北京地区云宏观特征进行统计分析。云出现率方面，4年平均值约36.3%；冬季最低，夏季最大；月出现率值9月最大，12月最小；出现率日变化有季节差异，春夏两季呈现中午（11:00，北京时间，下同）开始逐步升高至下午17:00后逐步下降的特点，增高幅度大于15%；冬、秋两季日变化特征不显著。高度方面，4年平均云底高约4.9 km，平均云顶高约7.2 km；云顶高和云底高的月变化特征明显，从年初1月开始逐步上升，在6月达到峰值，而后下降到12月达到低值；3～10月，高云（云底高>5 km)占约一半左右比例；厚度小于1 km的云在各月中所占比例最高；厚度1～4 km的云，厚度越大所占比例越低；特别地，厚度大于4 km的云所占比例在4～9月中仅次于厚度小于1 km云的比例。4年期间，北京地区单层云居多约占66.7%，两层云占比约25.2%，两层以上云占8.1%；冬季约80%的云为单层云，而6～9月云层分布变化最多，其中9月单层云比例最低约为40%。本文基于4年高时空分辨率雷达数据对北京地区云分布特征，特别是云垂直分布特征在数值上准确刻画，该项工作在已有云气候研究中尚未见开展，所获得的知识将对了解地区气候特征、区域模式云参数化选择提供参考。     

Observational Characteristics of Cloud Vertical Profiles over the Continent of East Asia from the CloudSat Data

The CloudSat satellite data from June 2006 to April 2011 are used to investigate the characteristics of cloud vertical profiles over East Asia(20°-50°N,80°-120°E),with particular emphasis on the profiles of precipitative clouds in comparison with those of nonprecipitative clouds,as well as the seasonal variations of these profiles.There are some obvious differences between the precipitative and nonprecipitative cloud profiles.Generally,precipitative clouds mainly locate below 8 km with radar reflectivity in the range of-20 to 15 dBZ and maximum values appearing within 2-4-km height,and the clouds usually reach the ground;while nonprecipitative clouds locate in the layers of 4-12 km with radar reflectivity between-28 and 0 dBZ and maximum values within 8-10-km height.There are also some differences among the liquid precipitative,solid precipitative,and possible drizzle precipitative cloud profiles.In precipitative clouds,radar reflectivity increases rapidly from 11 to 7 km in vertical,implying that condensation and collision-coalescence processes play a crucial role in the formation of large-size drops.The frequency distribution of temperature at-15℃ is consistent with the highest frequency of radar reflectivity in solid precipitative clouds,which suggests that the temperatures near-15℃ are conductive to deposition and accretion processes.The vertical profiles of liquid precipitative clouds show almost the same distributions in spring,summer,and autumn but with differences in winter at mainly lower levels.In contrast,the vertical profiles of solid precipitative clouds change from spring to winter with an alternate double and single high-frequency core,which is consistent with variations of the frequency distribution of temperature at-15℃.The vertical profiles of nonprecipitative clouds show a little change with season.The observations also show that the precipitation events over East Asia are mostly related to deep convective clouds and nimbostratus clouds.These results are expected to be useful for evaluation of weather and climate models and for improvement of microphysical parameterizations in numerical models.     

Assessment of FY-4A and Himawari-8 Cloud Top Height Retrievalthrough Comparison with Ground-Based MillimeterRadar at Sites in Tibet and Beijing

正1Key Laboratory of Middle Atmosphere and Global Environment Observation, Institute of Atmospheric Physics,Chinese Academy of Sciences, Beijing 100029, China2University of Chinese Academy of Sciences, Beijing 100049, China     

基于CloudSat资料对中国低纬度陆地区域卷云物理特征的研究

利用2008年9月—2016年8月的CloudSat卫星资料对发生在我国低纬度陆地区域(5°~36.5°N,78°~124°E)的卷云物理特征进行统计分析,并分别讨论东部沿海、中部、西部3个子区域的卷云物理特征的季节变化。结果表明:卷云的整层发生率西部地区整体低于中部与东部沿海地区。3个子区域整层发生率均在夏季最高、冬季最低。卷云的主要发生高度在5.04~18.71 km,垂直分布中卷云发生率的最大值出现在春季中部地区,为15.34%,高度为9.83 km。冰水路径最大值出现在夏季的东部沿海,液水路径最大值在秋季的西部地区。冰水含量、冰粒数浓度、冰粒有效半径的主要分布高度与卷云的发生高度一致,液水含量、液滴数浓度、液滴有效半径的主要分布高度在5.04~9.35 km。3个子区域卷云冰水含量、冰粒数浓度、冰粒有效半径垂直分布中大多集中在中上部;液水含量垂直分布主要集中在分布高度的中下部。四季卷云雷达反射率因子的最大值在-19.89~-16.78 dBZ,分布高度在7.19~10.55 km。     

Analysis of Ice Water Path Retrieval Errors Over Tropical Ocean

Retrieval of multi-layered cloud properties, especially ice water path (IWP), is one of the most perplexing problems in satellite cloud remote sensing. This paper develops a method for improving the IWP retrievals for ice-over-water overlapped cloud systems using Tropical Rainfall Measuring Mission (TRMM) Microwave Imager (TMI) and Visible and Infrared Scanner (VIRS) data. A combined microwave, visible and infrared algorithm is used to identify overlapped clouds and estimate IWP separately from liquid water path. The retrieval error of IWP is then evaluated by comparing the IWP to that retrieved from single-layer ice clouds surrounding the observed overlapping systems. The major IWP retrieval errors of overlapped clouds are primarily controlled by the errors in estimating the visible optical depth. Optical depths are overestimated by about 10–40% due to the influence of the underlying cloud. For the ice-over-warm-water cloud systems (cloud water temperature Tw > 273 K), the globally averaged IWP retrieval error is about 10%. This cloud type accounts for about 15% of all high-cloud overlapping cases. Ice-over-super-cooled water clouds are the predominant overlapped cloud system, accounting for 55% of the cases. Their global averaged error is 17.2%. The largest IWP retrieval error results when ice clouds occur over extremely super-cooled water clouds (Tw 6 255 K). Overall, roughly 33% of the VIRS IWP retrievals are overestimated due to the effects of the liquid water clouds beneath the cirrus clouds. To improve the accuracy of the IWP retrievals, correction models are developed and applied to all three types of overlapped clouds. The preliminary results indicate that the correction models reduce part of the retrieval error.     

渤海海效应暴雪云特征的观测分析

利用静止卫星 (GMS-5, GOES-9, MTSAT) 红外数据与CloudSat卫星云剖面雷达数据、NCEP FNL分析资料与常规观测资料,对2001—2010年发生的12次渤海海效应暴雪过程中云的演变特征、渤海热力作用与暴雪云团垂直结构及相态组成进行了观测分析。发现不同生成源地的暴雪云通常在渤海上快速发展,云中多存在水平范围可达100~300 km的密实条状或块状云团,其下对应主要降雪区域;暴雪云生成源地可分为渤海湾及莱州湾附近、渤海中部、辽东湾附近3种,暴雪云在海上移动主要受850 hPa风场影响;渤海暖海面与其上冷空气间的热量、水汽交换形成的不稳定层结条件,导致暴雪云进一步发展;暴雪云发展旺盛时期高度可达4 km,其冰水含量最大值达600 mg·m-3且主要集中在2 km高度附近,平均值可达303 mg·m-3,冰粒子有效半径最大值约为120 μm,平均值约为91 μm。     

双雷达风场反演拼图在登陆台风“莫兰蒂”(1614)强降水精细预报中的同化应用试验

利用福建龙岩、漳州、泉州新一代多普勒天气雷达和厦门海沧双偏振雷达探测资料,采用动态地球坐标系下双雷达三维风场反演与拼图技术,基于天气研究和预报模式（Weather Research and Forecasting,WRF）及其资料同化系统,对登陆台风“莫兰蒂”（1614）引起的2016年9月14—15日福建强降水过程进行了双雷达风场反演拼图资料检验及其三维变分同化对强降水精细预报影响的数值试验,结果发现:（1）动态地球坐标系下双雷达反演风场能合理反映实际风场分布状况,其误差相对较小。相较厦门翔安风廓线雷达及厦门探空秒级测风数据,反演风风向（风速）平均绝对误差分别为7.8°（2.6 m/s）及3.4°（1.1 m/s）;（2）反演风场水平方向稀疏化对同化及预报结果极为重要,过密的反演风场资料会给同化及预报结果带来负效果。文中采用18、6、2 km 3重嵌套,在3重嵌套区域均进行同化以及仅在2 km区域进行同化两种情况下,均表现为当反演风场资料水平分辨率提高到0.1°时,同化分析及预报的台风环流开始受到负影响;且当反演风场资料水平分辨率越高时,负效果越明显。敏感性试验结果显示,分辨率取0.2°时数值预报效果最好;（3）以美国国家环境预报中心全球预报系统（National Centers for Environmental Prediction/Global Forecast System,NCEP/GFS）0.5°×0.5°分析场为初值,基于3个不同起报时刻（2016年9月14日14时、20时及15日02时）（北京时,下同）模拟的福建省境内台风内核雨带和螺旋雨带逐时演变、台风路径与强度、逐时降水TS评分和空间相关差异显著,其中14日14时起报试验效果最好;而14日20时起报试验效果最差,这与该试验初始台风大风轴风速明显偏大有关;（4）在上述3个不同起报时刻试验基础上,分别增加双雷达反演风场资料的三维变分同化后,福建境内地面风场和台风内核雨带、螺旋雨带逐时分布、逐时降水TS评分和空间相关、台风环流结构以及U、V风垂直廓线分布均有明显改善,最大正影响时效可达24 h;但仅对1—6 h时效内台风路径有改善。      

APPLICATIONS OF THE CLOUDSAT TROPICAL CYCLONE PRODUCT IN ANALYZING THE VERTICAL STRUCTURE OF TROPICAL CYCLONES OVER THE WESTERN PACIFIC

Cloud profiling radar (CPR) onboard CloudSat allows for deep penetration into dense clouds/precipitation. In this study, tropical cyclones (TCs) are classified into three stages as developing, mature, and decaying. The circular TC area with the radius of 500 km is divided into five regions. The vertical structure characteristics of 94 Western Pacific TCs at different stages in different regions from June 2006 to February 2014 are statistically quantified using the CloudSat tropical cyclone overpass product (the CSTC Product). Contoured frequency by altitude diagrams (CFADs) of radar reflectivity show an arc-like feature and exhibit opposite distributions with a boundary at 5 km. Bright bands are found at this altitude, indicating melting layers. Deep convective (DC) clouds have the largest occurrence probability in the inner region, while Ci clouds occur more frequently in the outer region at 10-15 km. As clouds have the second largest vertical scale after DC clouds. Distributions of Ac, Cu, and Ns clouds at different stages have few distinctions. As the altitude increases, the ice effective radius and the distribution width parameter decrease while the particle number concentration increases. Moist static energy (MSE), cloud thickness (CT), liquid water path (LWP), ice water path (IWP), water vapor (WV), and rain rate (RR) all diminish along the radial direction and are significantly larger at the mature stage. The average value of MSE at the developing stage is larger than that at the decaying stage.     

催化对降水区域外效应的中尺度数值模拟

The Fifth-Generation NCAR/Penn State Mesoscale Model (MM5) has been used to investigate the extra-area effects of silver iodide (AgI) seeding on stratiform clouds performed at the supercooled layer.A bulk two-moment microphysical scheme and the new software package for silver iodide are incorporated in MM5.Extra conservation equations are applied to trace the seeding agent,which is transported along the flow field and interacts with the supercooled cloud fields.In this study,the model was run using three nested grids,with 3.3 km × 3.3 km horizontal resolution in the finest grid.The model results showed that seeding with AgI at the 5 to 15℃ levels had microphysical effects on the simulated clouds and that the simulation produced a longer-lasting seeding effect because of the transport of the seeding agent by upper-level winds.Most of the AgI particles acted as deposition nuclei,and the deposition nucleation process contributed mostly to additional cloud ice formation in this study.The results showed that more precipitation results from seeded than unseeded case,and the precipitation was redistributed downwind of the target.Augmented precipitation (varying from 5% to 25% downwind) was confined in space to within 250 km of the seeding target and in time to the 3-h period after initial seeding.