基于漂移克里金融合雷达、雨量计定量估测降水研究

文中介绍了一种新的融合雷达和雨量计数据开展定量估测降水研究的空间信息统计学方法—Kriging with external drift(KED)方法。该方法能很好地融合高精度、低时空分辨率的雨量计数据和低精度、高时空分辨率的雷达数据进行插值。通过变异函数描述降水场的空间结构信息,能够充分利用数据间的空间相关性,来改进估测精度和提高处理速度。利用其优良的数学特性,以期在定量估测降水业务研究上进行新的探索和尝试。选用湖南省有代表意义的3次降水过程资料,通过雷达直接估测降水（RAD）、变分校准（VAR）以及KED 3种方法,分别与雨量计测量值进行对比分析,选用代表站进行交叉验证结果均表明：RAD的均方差、绝对误差、相对误差最大,VAR次之,而KED最小。KED估测的结果与雨量计测量降水最为接近,估测效果最好;3种方法与雨量计实测值计算一定范围的误差频率,KED估测值具有最小的均方差和最小的标准差,且误差分布相对集中在0值附近,斜度和峰度最佳,试验证明该方法不仅能提高降水估测精度,且优于其他方法,VAR均方差次之,RAD均方差效果相对较差。联合雷达、雨量计估测降水的实质是把雷达估测值与雨量计测量的结果相融合,以雨量计来校准雷达估测值,保留了雷达探测到降水的中、小尺度精细特征。校准后的雨量场数值接近雨量计测值,而且能够准确反映雷达测得的降水分布形式。     

On the Tropical Rainfall Measuring Mission (TRMM)

Summary The importance of quantitative knowledge of tropical rainfall, its associated latent heating and variability is summarized in the context of climate change. Since the tropics are mainly covered with oceans, with some deserts and jungles, the monthly precipitation is not known within a factor of two. Hence the only way to measure it adequately for climate and general circulation models is from space. The paper describes the Tropical Rainfall Measuring Mission (TRMM). This joint Japan-U.S. cooperative Earth Probe satellite will be launched from Japan in 1997 for a three-year mission. The scientific basis of the instrument and orbit selection is explained. The precipitation instrument complement comprises the first rain radar to be flown in space (PR), and a multi-channel passive microwave sensor (TMI) improved relative to the SSM/I¹ by an additional channel at 10 GHz. The third rain instrument is a five-channel VIS/IR (VIRS) sensor. Progress in construction of instruments, observatory, data system, and the ground validation program is summarized. A report is also given concerning development of the algorithms by which rainfall and its associated latent heat release will be calculated from the several instruments, separately and in combination, and how the scientists will interact with the data system to obtain the 32 rain data products necessary to fulfill the science requirements.With 9 FiguresSSM/I stands for Special Sensor Microwave Imager.     

微波链路测量降水研究综述

利用微波链路测量降水是降水测量的最新技术,具有作用于近地面大气、时空分辨率高等优势。文章总结了微波链路测量降水方法的发展现状和技术难点,结合微波通信的实际应用,分析了频段的选择依据、雨衰减量的精确测量和正演模型的精确建模需要考虑的因素,指出降水起始时间的准确识别和区域化反演算法是解决该方法实用化问题的关键技术,为利用微波链路测量降水的进一步研究提供了借鉴。     

利用雷达资料对自动雨量计实时质量控制的方法研究

自动雨量计资料是对降水的直接测量,在流域面雨量计算、气候研究、气象服务等方面具有重要意义。但是,由于风力、蒸发、灌溉、校准、漏斗堵塞、机械故障、信号传输等原因往往造成其存在不同类型的系统误差和随机误差, 自动雨量计数据在定量使用前需要进行质量控制。目前,天气雷达以其高时空分辨率的优势已经成为监测降水的重要手段,本文首先采用两步校准法改善雷达估测降水,然后对雷达—雨量计对之间的差异进行统计学的分析,确定自动雨量计质量控制的一些标准,从而对雨量计进行质量控制。最后用两个降水过程对自动雨量计质量控制的结果进行了检验,结果表明:两步校准法改善了雷达估测降水的系统性偏差,并减小了雨量计站点上的相对误差;可以利用雷达估测降水实现对自动雨量计的实时质量控制,就整个数据集而言,约0.1%的数据被怀疑为误判,误判的自动雨量计主要位于雨带的边缘。但该质量控制算法同时也存在一定的局限性:在雨带的边缘或没有天气雷达覆盖的区域,以及雷达资料存在数据质量问题的情况下,往往会造成对雨量计的误判。     

基于雷达—雨量计降水融合方法提高极端降水监测能力

高时空分辨率、高精度的降水产品对于极端降水的监测以及防灾减灾具有重要意义。地面雨量计提供点尺度降水精确观测,但无法精细化捕捉对流性强降水的空间分布。雷达观测可以精细地刻画降水的空间分布特征,但雷达定量估计降水（QPE,quantitative precipitation estimation）产品估测精度易受雷达观测偏差和Z–R（雷达反射率—降水率）关系等因素影响。因此,本文开展高时空分辨率的雷达—雨量计降水融合算法研究,集成雨量计观测和雷达定量估计降水产品各自的优点。该算法主要步骤包括:雨量站观测数据格点化、局地雨量计订正雷达QPE和雷达—雨量计降水融合三个部分。首先利用克里金插值方法,对雨量站观测的降水进行插值,得到格点降水信息;再通过局地雨量计订正方法系统性地订正雷达QPE产品,以提高雷达QPE产品精度;最后,结合降水类型,通过雷达—雨量计降水融合算法,产生高时空分辨率、高精度的雷达—雨量计降水融合产品。通过郑州“21·7”暴雨、台风“烟花”和2021年8月随州暴雨三个典型的极端降水个例,对雷达—雨量计降水融合算法产生的雷达—雨量计降水融合产品进行了系统地评估和分析。结果表明,在不同的极端降水个例和不同的降水时段,雷达—雨量计降水融合产品精度上优于雷达QPE产品,且在降水的空间分布上较雨量站观测格点插值产品更能精细地刻画降水的结构特征。表明算法得到的雷达—雨量计降水融合产品的准确性较高,对极端降水有较好地捕捉和监测能力。     

基于准同雨团样本概念雷达和雨量计的实时同步结合方法

文中提出一种基于准同雨团样本概念雷达与雨量计(R&G)的实时同步结合技术.由于存在探测时空和分辨率的两种差异,雷达与雨量计(R&G)的同样本对应历来被视为一道难题.但观测表明,雨量计与其垂直上空雷达的小时累积量(ZOH,QG)存在符合幂律的相关性.以此为基础,提出旨在基本消除时空差异的准同雨团样本概念和5个R&G直接对应关系,着重研究一种简便实用的"时间积分垂直同步采样"TIVS,以及具有幂律形式的ZOH-QG关系.有意义的是在固定指数条件下系数AB或AM,具有Z-R转换和雨量计调整(订正)同步结合于一式的功能,两种设备在探测分辨率上巨大差异而引起的降水估算误差由此得到明显缓解.这种利用ZOH-QG关系估算地面小时降水累积量的方法称为RASIM.经过两个降水个例的试验初步表明,雷达估算半径为230 km内地面区域小时累积降水量的正确率在90%以上,而全过程以站为计的点评估相对误差率平均为20%左右.本文还对时间积分垂直同步采样在具有风漂移效应各种环境风场中的可利用性作了详细分析,揭示了时间积分垂直同步采样方法的物理本质,明确指出这是一种近似的准同雨团样本采样;同时通过分析R&G数据对的分布规律,研究提出了有效的质量控制方法,明显改善了ZOH-QG关系的稳定性和合理性.     

临近预报的两种高时空分辨率定量降水预报融合算法的对比试验

长期以来,雷达回波外推技术是0—2 h临近预报系统主要采用的方法,但其实际有效预报时间≤1 h,而中尺度数值模式预报则受平衡约束时间的限制,最初2 h的降水预报无效。为解决上述两种预报的缺陷,目前国际上流行采用外推预报与数值模式预报融合的技术,形成统一的0—6 h格点化的高分辨率无缝隙定量降水临近预报系统。对目前流行的两种融合算法（INCA（Integrated Nowcasting and Comprehensive Analysis System）算法及RAPIDS（Rainstorm Analysis and Prediction Integrated Data-processing System）算法）进行了分析和对比试验,以期为业务应用提供借鉴。RAPIDS算法的核心是用自动气象站雨量融合雷达估测得到的定量降水对模式预报的降水强度和位相进行修正;INCA算法则是用数值模式预报的风场修正外推技术的降水移动矢量。两种方法在0—6 h预报时效内,外推预报的权重均逐渐减小,模式预报的权重逐渐增大,从而实现外推预报和模式预报的平滑过渡。试验结果表明,两种方法对降水雨带和降水强度的预报均优于单一的外推预报或模式预报。集二者的优势研发最优的高时、空分辨率降水预报无缝隙融合算法,将有助于进一步提升高分辨率定量降水0—6 h无缝隙预报水平。      

漂移克里金方法在雷达和雨量计联合估测降水中的应用

文中介绍了一种新的融合雷达和雨量计数据开展定量估测降水研究的空间信息统计学方法-Kriging with externaldrift(KED)方法.该方法能很好地融合高精度、低时空分辨率的雨量计数据和低精度、高时空分辨率的雷达数据进行插值.通过变异函数描述降水场的空间结构信息,能够充分利用数据间的空间相关性,来改进估测精度和提高处理速度.利用其优良的数学特性,以期在定量估测降水业务研究上进行新的探索和尝试.选用湖南省有代表意义的3次降水过程资料,通过雷达直接估测降水(RAD)、变分校准(VAR)以及KED3种方法,分别与雨量计测量值进行对比分析,选用代表站进行交叉验证结果均表明:RAD的均方差、绝对误差、相对误差最大,VAR次之,而KED最小.KED估测的结果与雨量计测量降水最为接近,估测效果最好;3种方法与雨量计实测值计算一定范围的误差频率,KED估测值具有最小的均方差和最小的标准差,且误差分布相对集中在0值附近,斜度和峰度最佳,试验证明该方法不仅能提高降水估测精度,且优于其他方法,VAR均方差次之,RAD均方差效果相对较差.联合雷达、雨量计估测降水的实质是把雷达估测值与雨量计测量的结果相融合,以雨量计来校准雷达估测值,保留了雷达探测剑降水的中、小尺度精细特征.校准后的雨量场数值接近雨量计测值,而且能够准确反映雷达测得的降水分布形式.     

星载双频风场雷达热带气旋降雨区测风模拟

基于大气辐射传输理论分别建立Ku波段和C波段的降雨模型,模拟热带气旋降雨区洋面的雷达回波并反演了洋面10 m风场,用于研究降雨对测风的影响以及风云三号双频风场雷达 (WFR) 的测风能力。分析表明:回波的衰减或增强取决于降雨衰减项和后向散射项的相对大小;热带气旋的降雨使反演风速偏小,风向精度降低,Ku波段相对于C波段更易受影响,在高风速 (超过30 m·s-1) 条件下,可达5~20 m·s-1的负风速偏差。反演结果表明:双频反演的新方法能够结合Ku波段与C波段的优势,双频最大似然估计 (MLE) 方法在分辨率上优于C波段单频反演,相对Ku单频反演能降低降雨对测风的衰减作用,结合双频MLE方法和C波段单频反演优势的分区反演方法可以显著减小降雨偏差,提高风速反演精度,在有风云三号湿度计同步观测的条件下,是提高热带气旋降雨区测风精度的有效手段。     

微波辐射计与探空仪测值对比分析

利用武汉站高时空分辨力探空资料(3h,30m)与同址的MP-3000A型地基微波辐射计资料,分析了微波辐射计探测偏差的日变化及时间序列变化特征。结果表明,微波辐射计对大气要素的探测具有较高的探测准确度,但仍然存在一定的探测偏差。温度的探测偏差大值区出现在午后,相对湿度及水汽密度的探测偏差大值区则在凌晨;温度及相对湿度探测误差的时间序列表现为起伏较大,而水汽密度则相对平稳,降水,特别是强降水,会对微波辐射计的探测准确度产生较明显的影响。     

新一代天气雷达定量降水估测集成系统

降水的定量测量是天气雷达的重要应用之一,新一代天气雷达定量降水估测集成系统(QPEGS)是一套基于新一代天气雷达的雷达雨量计联合估测降水软件,利用多种雨量计校准雷达降水的方法,生成1小时降水分布。其产品的时间分辨率达10分钟,空间分辨率1 km×1 km,适合省市级业务台站使用。过去3年的数据评估表明:校准雨量计数量和估测精度有明显正相关,校准雨量计数越多,降水估测精度越高,2003年的小时降水估测误差约40%,过程降水量的估测误差小于20%;雨量计密度保持不变的情况下,降水时段越长,降水区域越大,降水估测的精度也越高。     

C波段双偏振雷达降水估计的误差分析与建模

双线偏振雷达定量降水估计精度受多种因素影响,为了更好地应用双偏振雷达估计降水并进一步提高降雨估测精度,需对雷达降水估计进行误差分析和建模.基于2015—2016年南京信息工程大学C波段双偏振雷达、雨滴谱仪观测资料以及南京地区雨量计数据,统计分析雷达估测降水的误差分布,分离雨量计代表性误差,并对随机误差和系统误差量化建模...     

用TRMM资料研究江淮、华南降水的微波特性

热带测雨卫星TRMM (Tropical Rainfall Measuring Mission) 于1997年11月发射成功, 其首次携带了空载雷达, 有关资料已在网上对公众发布。利用热带测雨卫星上的微波成像仪TMI (TRMM Microw ave Imager) 资料以及其和测雨雷达TRMM/PR (Precipitation Radar) 资料联合反演的地面瞬时降水产品, 采用散射指数 (Is) 法从理论上探讨了我国江淮、华南降水尤其是暴雨的微波特性, 其中I_s表达式通过江淮、华南晴空TMI资料统计回归得到。以联合反演的地面瞬时降水产品为真值, 用面积相当法对14个降水个例求I_s降水阈值, 研究了阈值和降水面积以及85.5 GHz垂直通道最低亮温的关系, 并寻求了I_s和降水的相关特征。研究表明:I_s降水阈值随降水面积的增大或85.5 GHz垂直通道最低亮温的降低有增高的趋势; I_s与强对流性降水瞬时雨强对应很好, I_s≥60 K是一个好的暴雨指标。最后进行了初步的雨强反演试验研究, 由于TMI资料分辨率的提高以及时空配合较好的真值, 反演的地面瞬时降水与真值相关效果大大提高。     

Parsivel~激光降水粒子谱仪及其在气象领域的应用

降水粒子特性是大气运动和云内微物理过程的综合结果,在云降水物理及人工影响天气领域有着重要的意义。传统的测量方法不适合对大量数据分析寻找规律,德国OTT公司的Parsivel激光降水粒子谱测量系统可以较好解决自动测量难题。该仪器是以激光测量为基础的粒子测量传感器,采用平行激光束和光电管阵列结合,当有降水粒子穿越采样空间时,自动记录遮挡物的宽度,通过穿越时间计算降水粒子的尺度和速度,根据各种参数的综合信息对降水粒子进行分类,并能够以数字形式显示瞬时降水强度、降水粒子总数、累积降水量、降水时的能见度和雷达反射因子,以图形方式显示降水粒子尺度谱、速度谱、降水粒子分类且自动生成天气现象代码,实现天气现象的自动识别。激光降水粒子谱仪主要用于气象水文观测。在雷达气象学领域可用于Z/R关系的拟合修正,比传统的用雨量筒观测数据拟合效果好得多;由降水粒子谱仪测量雨滴的降落速度,可以对天气雷达垂直向上测量的粒子径向速度谱进行校正。人工影响天气的效果检验一直是一个难题,自然降水粒子谱分布形式与人工催化以后的降水粒子谱型理论上应当具有较大的差别,人工增雨作业降水滴谱变化物理响应和降水强度时间变化响应都有明显的区别。如果能够实时检测到这些差别,就能够充分说明人工催化的有效性。未来如果能够进行联网观测记录区域性降水、降雹,就有充分证据表明人影作业的有效性,在定量化作业效果评估以及灾害损失评估等方面应用潜力巨大。利用该仪器已经对一年的自然降水过程进行了连续观测,并将所获得的降水粒子谱、雨滴浓度值随时间变化状况与卫星反演的云顶有效粒子半径时间变化趋势进行了对比,发现有较好的一致性。     

The analysis of results of remote sensing monitoring of the temperature profile in lower atmosphere in Bergen (Norway)

Considered is the application of MTP-5 meteorological temperature profiler used for the remote sensing of vertical profiles of the air temperature in the planetary boundary layer and the lower one-kilometer layer of the atmosphere. The measurements were carried out in Bergen (Norway) in 2011–2012. The obtained dataset of temperature profiles has temporal resolution of five minutes and vertical resolution of 50 m. The MTP-5 data are complemented with the measurements of the air temperature and the wind taken at two automatic weather stations and with the measurements of the rain intensity made with the rain radar. Studied is the impact of meteorological conditions and precipitation on the MTP-5 readings. It is revealed that formation of a thin water film (of ice or, to a smaller degree, of sleet) on the surface of the sensor cover of MTP-5 has a significant impact on the data of the temperature monitoring. The removal of intensive precipitation (the precipitation rate is >0.2 mm/hour) improved the reliability and quality of the temperature profile monitoring. In particular, it is demonstrated that significant air pollution and stably stratified atmospheric conditions which lead to low temperatures are reliably monitored with this instrument.     

Passive microwave and infrared structure of mesoscale convective systems

Summary The precipitation structure of mature Mesoscale Convective Systems (MCS) is examined in both the midlatitudes and the tropics using SSM/I microwave measurements, geostationary satellite observations, and ground-based radar observations. Discussion includes qualitative comparisons between midlatitude and tropical MCS cases, with particular emphasis on the delineation of convective and stratiform regions and the characterization of microwave polarization difference temperatures in the MCSs. Implications are given regarding the importance of the vertical precipitation structure on top of the atmosphere (TOA) microwave temperatures and for rain retrieval algorithms using measurements from space.Some of the principle findings include the ability of passive microwave brightness temperature measurements to distinguish stratiform and convective regions of MCSs for both tropical and midlatitude cases and over land and ocean backgrounds. Convective regions typically had low differences between the vertical and horizontal brightness temperatures while the stratiform regions have larger differences, and these differences are likely related to the spatial microphysical variations in the upper levels of the precipitation region. Several cases were found in midlatitudes and one case in the tropics where the lowest infrared (IR) brightness temperatures were displaced into the anvil region and were not colocated with the coldest microwave temperatures. Life cycle dependence of the displacement is suggested, but the SSM/I measurements with a maximum of twice daily coverage over the same location were inadequate to answer this question.With 10 Figures     

Real-Time Synchronous Integration of Radar and Raingauge Measurements Based on the Quasi Same-Rain-Volume Sampling

A technique for real-time synchronous integration of radar and raingauge measurements based on the concept of the quasi same-rain-volume sampling(QSVS)is presented.Because of the temporal and spatial discrepancies and resolution differences,the integration of radar measurements with raingauge observations has long been a difficult task.Observations indicate that there exists a correlation that conforms to the power law between hourly accumulated raingauge measurement(QG)and detected radar echo(ZOH)over the raingauge.On the basis of this,a concept of the QSVS and five direct correspondent formulas of radar and raingauge samples are built up,aiming to eliminate the temporal and spatial discrepancies.A convenient and practical sampling method-the time integral vertical synchronous sampling(TIVS)is proposed and the ZOH-QG relationship is studied.It is significant that under the fixed exponent,the coefficient AB or AM varies flexibly in accordance with the temporal and spatial variability of natural precipitation,having the function of synchronously integrating the Z-R conversion and the gauge adjustment into a single equation,and thus the precipitation estimation errors caused by detecting resolution differences between radar and raingauge can be obviously mitigated.The real-time synchronous integration technique using the ZOH-QG relationship to estimate the ground hourly rainfall accumulation is called the radar-gauge synchronous integration method(RASIM).The experiments of two cases show that the accuracy of estimated surface hourly rainfall accumulation within 230 km is about 90%,and the average relative error for the point estimation over the whole process is about 20%.Through the detailed analysis of the applicability of TIVS in three environmental fields with various wind drifts,the physical essence of TIVS is explored: it is an approximate QSVS.By analyzing the data pairs of radar and raingauge,an effective quality-control procedure is established,which can greatly improve the stability and rationarity of the ZOH-QG relationship.The forecasting product of hourly rainfall accumulation derived from the RASIM has been put into operation.It is demonstrated that the RASIM plays an important role in the quantitative monitoring and forecasting of short-term torrential rainfall.     

中国区域高分辨率多源降水观测产品的融合方法试验

高质量、高分辨率降水产品研制对于数值天气模式检验、水文陆面模拟、山洪地质灾害监测有着重要意义。利用中国近4万自动气象站逐时降水资料、中国雷达定量降水估计和CMORPH卫星反演降水产品,开展0.05°×0.05°和0.01°×0.01°两种高分辨率下的三源降水融合方法研究试验,探讨如何有效引入雷达高分辨率信息来提高降水产品质量。一方面,在0.05°分辨率上,先以自动气象站观测降水数据为基准,采用概率密度函数（PDF）匹配法订正雷达和卫星估测降水产品的系统偏差,将雷达降水产品的偏差从-0.05 mm/h降至-0.008 mm/h;再采用贝叶斯模型平均（BMA）方法融合雷达和卫星降水产品,形成0.05°分辨率的中国区域覆盖完整且最优的联合降水背景场。此外,在0.01°分辨率上,以0.05°分辨率的卫星-雷达贝叶斯模型平均联合降水产品为背景,采用1 km雷达估测降水的空间结构信息进行降尺度,亦能有效提高0.01°分辨率背景场的质量。然后,分别以不同分辨率的卫星-雷达联合降水产品为背景,采用统计方法量化误差估计,再采用最优插值方法融入地面观测。通过2419个中国国家级气象台站的独立样本检验,评估了多种类型的降水资料及融合试验产品在中国地区的质量。结果表明,两种分辨率的三源融合试验产品的精度均优于任何单一来源的降水产品,特别是在站点稀疏地区,降水精度均较融合前有显著提高,达到了较好的融合效果,其中在0.05°分辨率上采用“概率密度函数+贝叶斯模型平均+最优插值”方法的三源融合降水产品整体质量最好,而0.01°分辨率上基于“概率密度函数+贝叶斯模型平均+降尺度+最优插值”方法的三源融合降水产品在强降水监测上更有优势。      

Precipitation in the Canadian Atlantic storms program: Measurements of the acoustic signature

Abstract

The Canadian Atlantic Storms Program (CASP) provided an opportunity for comparing two quite different remote‐sensing approaches to the detection of precipitation: radar backscatter and ambient ocean sound. Several of the gales passing the observation area during CASP produced substantial precipitation with the periods of radar backscatter showing close coincidence with simultaneous acoustic signals. The ambient sound record most readily yields an indication of precipitation from shifts in spectral slope. An important result of the experiment is the demonstration that an identifiable precipitation signal occurs even in strong wind conditions. The surface bubble layer formed during strong winds only partially attenuates the higher frequency acoustic components generated by precipitation. During rain‐free periods the attenuation can be interpreted in terms of the bubble size distribution.     

雷达与雨量计联合估测降水的相关性分析

在对比分析质量控制前后雷达估测降水量与自动雨量计降水量之间相关性的基础上,采用雷达-雨量计联合校准方法,对14种不同密度雨量计校准雷达估测降水的效果进行分析。结果表明:在使用雷达资料和雨量计资料前有必要对资料的质量进行分析与控制。联合雨量计校准雷达能明显提高雷达对降水的估测能力;采用不同密度雨量计校准雷达,随着校准雨量计密度的加大,雷达估测降水的精度不断提高并趋于稳定。校准雷达的效果及所需雨量计密度与降水类型有关,当校准效果相同时,积云强降水过程需要的雨量计密度最大,积混对流性降水过程次之,层云稳定性降水过程需要的雨量计密度最小。不同方法的校准效果不同,卡尔曼滤波方法适合于对稳定性降水的校准,或在雨量计密度低的地区对雷达进行校准;变分校准法和最优插值法的校准效果相当,适合对积混对流性降水的校准,或在雨量计密度高的地区对雷达进行校准。