Constraint inversion Algorithm of Lidar Equationfor Deriving Aerosol Optical Property

1.IntroductionLidarisincreasinglyusedinmeasuringtheopticalpropertiesofaerosolsandcloudssincethe1960's.Ithasbeenshownthatthedeterminationoftheextinctioncoefficientasafunctionofheightisfeasible(Klett,1980;Fernald,1984;LuDarenetal.,1977;Spinhirneetal.,...     

Two-wavelength Lidar Measurement of Cloud-aerosol Optical Properties

Ｔｗｏ－ｗａｖｅｌｅｎｇｔｈＬｉｄａｒＭｅａｓｕｒｅｍｅｎｔｏｆＣｌｏｕｄ－ａｅｒｏｓｏｌＯｐｔｉｃａｌＰｒｏｐｅｒｔｉｅｓＱｉｕＪｉｎｈｕａｎ（邱金桓）（ＩｎｓｈｔｕｔｅｏｆＡｔｍｏｓｐｈｅｒｉｃＰｈｙｓｉｃｓ，ＣｈｉｎｅｓｅＡｃａｄｅｍｙｏｆＳｃ...     

Constraint Inversion Algorithm of lidar equation for deriving aerosol optical property

A key question of the backward integration algorithm to lidar equation is how to determine the far-end boundary value. This paper develops a Constraint Inversion Algorithm (CIA) for deriving the value and then the aerosol extinction profile from lidar signals, which uses the ground-level horizontal lidar signals as the constraint information. The smaller the wavelength is, the more sensitive to the variation of aerosol ex-tinction to backscatter ratio solved by CIA. According to the property an algorithm is further proposed to simultaneously retrieve the aerosol extinction profile, the size distribution and the imaginary part of its re-flective index from the multi-wavelength lidar observations. CIA is tested in the inversion simulations with satisfactory result.     

A New Method for Aerosol Retrieval Based on Lidar Observations in Beijing

Lidar has been used extensively in the area of atmospheric aerosol measurement. Two unknowns at the reference altitude, the lidar ratio and the backscatter coefficient, need to be resolved from the lidar equation. In the actual application, these two values are difficult to obtain, particularly the backscatter coefficient. To better characterize the optical properties of aerosols, optical thickness, and attenuated backscatter obtained by other instruments are usually used as the input for joint inversion. However, this method is limited by location and time. In this study, the authors propose a new method for aerosol retrieval by using Mie scatter- ing lidar data to solve this problem. The authors take the horizontal aerosol extinction coefficient as the con- straint to begin the iteration until a self-consistent aerosol vertical profile was obtained. By comparing their results with Aerosol Robotic Network （AERONET） data, the authours determine that the aerosol extinction coefficient obtained by combining horizontal and vertical lidar observations is more pre- cise than that obtained by using the traditional Fernald method. This new method has been adopted for re- trieving the extinction coefficient of aerosols during the observation days.     

参数化的多次散射雷达方程和它的反演理论 Ⅱ:反演理论

本文根据参数化的多次散射雷达方程,分析了激光多次散射对激光回波信号和一次散射雷达方程反演结果的效应,提出了迭代前向积分法和迭代后向积分法以求解参数化的多次散射雷达方程,并根据多次散射信号对云和气溶胶的前向散射相函数的高度敏感性提出了一个从双接收视场的多次散射激光回波信号,同时确定云和气溶胶消光系数分布和前向散射相函数的方法.此外,本文还根据数值试验分析了这些反演方法在地基激光雷达和空间站激光雷达两种情形下的适用性.     

北京秋季一次降雪前污染天气的激光雷达观测研究

以2009年11月5~8日北京地区发生的一次特殊天气形势下的重污染天气过程为例,研究分析本次污染特点和大气边界层结构特征以及此天气过程的大气温度和相对湿度结构特点。激光雷达是探测大气边界层及气溶胶的一个高效工具,利用ALS300激光雷达系统测量信号,应用Fernald方法反演大气消光系数,根据反演的气溶胶消光系数的最大突变,即最大递减率的高度来确定大气边界层的高度。利用其观测的退偏比分析大气污染物特性。利用微波辐射计数据,确定大气温度和湿度时空特征。研究结果表明:在本次污染天气下,大气具有很强的逆温结构,逆温最大可达近1 K(100 m)-1,500 m以上的大气相对湿度很低,在这种天气特征下的大气边界层高度在400 m左右,非常稳定。污染结束降雪开始前,大气逆温结构消失,大气湿度大幅度增加,接近饱和。根据lidar(light detection and ranging)退偏比的分析,本次污染天气是一次典型的烟尘类颗粒物的污染,污染具有区域性特点。PM2.5(空气动力学当量直径小于等于2.5μm的颗粒物)与AOT(Aerosol Optical Thickness)之间有明显的线性关系,相关系数达到0.72。该lidar系统能够反演出秋季降雪前本次污染天气背景下北京城区上空的大气污染特性和大气边界层高度。     

Sensitivity of lidar equation solution to boundary values and determination of the values

An analytical dependence of the optical depth solution to lidar equation on boundary values was con-firmed. According to the dependence this paper analyzed the sensitivity of lidar equation solutions obtained by forward and backward integration algorithms to the boundary values and quantitatively expounded an error limit to the boundary values under a given inversion accuracy. Furthermore, this paper presented a method for determination of the far-end boundary value in the case of inhomogeneous atmosphere, improving the accuracy of lidar equation solution.     

Extinction versus backscatter relationships for lidar applications at 351 nm: maritime and desert aerosol simulations and comparison with observations

Functional relationships linking at λ₀=351 nm aerosol extinction α_λ0^aer and backscatter coefficient β_λ0^aer of maritime and desert type aerosols are determined to allow for inversion of the single-wavelength lidar signals. Such relationships are derived as mean behavior of 20,000 extinction versus backscatter computations, performed for aerosol size distributions and compositions whose describing parameters are randomly chosen within the naturally observed variability. For desert-type aerosols, the effect of the particle non-sphericity is considered and it is shown that the extinction to backscatter ratio of non-spherical dust particles can be up to 60% larger than the values obtained for spherical particles. Aerosol extinction and backscatter coefficient profiles obtained inverting the single-wavelength lidar signal with the modeled relationships are then compared to the same profiles measured by a combined elastic-Raman lidar operating at 351 nm. Analytical back trajectories and satellite images are used to characterize advection patterns during lidar measurements and to properly choose the modeled functional relationship. A good accordance between the two techniques is found for advection patterns over the lidar site typical of maritime and dust conditions. Maximum differences between the model-based α_λ0^aer and β_λ0^aer vertical profiles and the corresponding ones measured by the combined elastic-Raman lidar technique are of 30% and 40% in maritime and desert dust conditions, respectively. The comparison of elastic-Raman lidar measurements and model-based results also reveals that particle non-sphericity must be taken into account when mineral dust-type aerosols are directly advected over the measurement site.     

双差分吸收激光雷达:-种能有效减小气溶胶对臭氧测量影响的方法

文中提出了一种新的激光雷达测量臭氧的方法：双差分吸收方法。理论分析和数值模拟表明这种方法可以有效减小气溶胶消光和后向散射对臭氧测量的影响，从而使激光雷达在气溶胶影响严重地区测量的臭氧精度比传统差分吸收激光雷达大大提高。利用（２８９，３１３；２７７．１，２９９．１ｎｍ）或（２６８．４，２８９；２７７．１，２９９．１ｎｍ）４波长进行双差分吸收可以用于对流层大气气溶胶含量丰富或分布不均匀地区臭氧的测量。利用（２９９．１，３４１．５；３０８，３５３ｎｍ）４波长进行双差分吸收可以对火山爆发后平流层臭氧进行较精确的测量。     

Study of the moist Convective Boundary Layer structure by backscattering lidar

Measurements of the structure of the Convective Boundary Layer (CBL) were made from the ground and from the ARAT aircraft with a backscattering lidar during the preliminary phase of the French project LEANDRE. A numerical model of Mie theory diffusion was used to determine a theoretical profile of the extinction coefficient, as a function of relative humidity. Comparisons between theoretical and measured profiles were made. Good agreement was found, which confirms that the variation of the extinction coefficient in the CBL is primarily controlled by the relative humidity. This paper also presents a detailed analysis of the small-scale structure of the CBL. The normalized length and number of thermals are estimated. Results are in agreement with previousin-situ measurements. An analysis of the bidimensional cross-section of the extinction coefficient shows that this coefficient permits one to retrieve more quantitative information than the range-corrected signal, as the fractions of entrained air and surface-layer air at a given altitude.     

拉曼雷达重叠函数的模拟与校准

重叠函数的校准对地基激光雷达低空大气探测的准确性至关重要。目前被广泛用于校准拉曼激光雷达重叠函数的双通道实验标定法,需要满足弹性散射通道与氮气拉曼散射通道的重叠函数近似相等的前置条件。但实际仪器光路往往会偏离理想状态,使得该前置条件无法得到满足,导致校准失败。本文使用光线追踪法模拟重叠函数,计算了各种光路失调情况下弹性散射通道与氮气拉曼散射通道的比值,并引入了一种镜头遮蔽实验来评估双通道实验标定法的前置条件是否得到满足。数值模拟结果表明,当雷达接收面的不同象限被遮蔽时,如果弹性散射通道与氮气通道的信号强度比值基本保持不变,则满足前置条件;如果弹性散射通道与氮气通道的信号强度比值有显著差异,则可判定仪器光学系统失调,需要调整直至满足上述前置条件时才能使用双通道实验标定法标定。利用该方法对南京市气象局安装的拉曼激光雷达进行了光路测评和调整,并与CCD(charge-coupled device)侧向成像激光雷达观测信号做对比,结果显示调整后的定标效果更好。     

Characteristics of cirrus clouds and its radiative properties based on lidar observation over Chung-Li, Taiwan

In this paper, characterization of cirrus clouds are made by using data from ground based polarization lidar and radiosonde measurements over Chung-Li (24.58°N, 121.10°E), Taiwan for a period of 1999–2006. During this period, the occurrence of cirrus clouds is about 37% of the total measurement nights over the lidar site. Analysis of the measurement gives the statistical characteristics about the macrophysical properties such as occurrence height, ambient temperature, and its geometrical thickness while the microphysical properties are interpreted in terms of extinction coefficient, optical depth, effective lidar ratio and depolarization ratio. The effective lidar ratio has been retrieved by using the simulation technique of backscattered lidar signals. The effect of multiple scattering has been taken into the account by a model calculation. Summer (Jun–Aug) shows the maximum appearances of cirrus due to its formation mechanism. It is shown that tropopause cirrus clouds may occur with a probability of about 24%. These clouds are usually optically thin and having laminar in structure with some cases resembling the characteristics similar to that of polar stratospheric clouds (PSCs). The radiative properties of the cirrus clouds are also discussed in detail by the empirical equations with results show a positive feedback on any climate change.     

参数化的多次散射激光雷达方程和它的反演理论 Ⅰ:方程

本文提出了一个参数化的多次散射激光雷达方程,该方程基于本文引出的四个因子,即几何消光因子、消光分布因子、前向散射因子和后向散射因子.前两个因子表征多次散射成分对大气消光系数分布、激光的发散角和接收视场角以及接收孔径的依赖关系,后两个因子表征多次散射成分对散射相函数的依赖关系.这个参数化多次散射雷达方程在241个数值试验中得以检验,这些试验包含很宽的大气条件和雷达几何参数,包括14个大气散射相函数,均匀和不均匀的大气消光系数分布,0.5至1之间变化的一次散射反照率,地基和空间站激光雷达两种情形.数值试验表明,在小于4的光学厚度内参数化的解和Monte-Carlo解之间的标准偏差小于27％,而本模式的计算时间比Monte-Carlo方法偏小4个数量级左右.本模式不仅适合于研究多次散射对激光回波信号和激光大气遥感的效应,而且对于考虑多次散射的激光大气探测而言,是一个合适的应用模式.     

合肥市郊低层大气的激光雷达探测研究

利用L300米散射激光雷达对合肥市郊大气边界层进行长期系统观测.分析讨论了大气边界层气溶胶消光系数与温度、湿度的关系, 大气边界层气溶胶消光系数垂直分布和时间变化的主要特征, 给出了激光雷达探测的大气边界层高度的统计特征及其与无线电气象探空仪探测大气边界层高度的比较结果.     

基于数据窗口标准差的边界层高度反演方法——以上海市为例

大气边界层具备一个重要特性,即在边界层顶部,气溶胶浓度在垂直分布上存在显著突变.利用该特性,采用主动遥感装置经过窗口平滑去除噪点后数据,提出基于梯度法改进的大气边界层高度反演方法——窗口标准偏差法.基于激光云高仪后向散射廓线数据,利用该方法反演边界层高度,在边界层高度下大气气溶胶混合均匀时,反演结果较为理想.在此基础上...     

气溶胶激光雷达和无线电探空观测边界层高度的对比分析

利用北京地区2017年11月至2018年1月连续3个月的激光雷达资料和无线电探空数据,按照清洁天、污染天和多云天3种天气条件,对大气边界层高度的计算方法和结果进行对比分析。结果表明,基于激光雷达消光系数的梯度法、标准差法和小波法都能够较好地提取边界层高度。清洁天标准差法计算的边界层高度高于梯度法和小波法,08:00（当地时间,下同）和20:00由无线电探空得到的清洁天边界层高度平均值分别为1176 m和1224 m。污染天标准差法的计算结果要低于梯度法和小波法,污染天无线电探空得到的边界层高度平均值约为956 m,和清洁天相比降低了两百多米,重污染时最低降低至562 m,逆温层高度和PM2.5浓度具有明显的反相关关系。有云时,梯度法和小波法确定的边界层高度和云高非常接近,标准差法计算的结果略低。总体而言,气溶胶激光雷达计算的边界层高度随着污染等级的提高没有明显的降低趋势,相反在重度污染情况下反而有所增加,这可能是由于污染物的不断堆积导致的。梯度法确定的边界层高度易受到污染物传输过程的影响,略高于逆温层高度。另外,激光雷达确定的边界层高度受到残留层影响时,也会高于逆温层。     

高光谱分辨率激光雷达同时测量大气风和气溶胶光学性质的模拟研究

提出一套高光谱分辨率激光雷达(HSRL)系统,用于同时测量大气风和气溶胶的光学性质.该HSRL系统中使用碘分子滤波器分离分子和气溶胶后向散射,同时利用双边缘检测技术测量大气风场引起的多普勒频移.文中选用合理的HSRL参数和大气模型数据,模拟和分析了HSRL的测量性能.系统夜晚运行时,可测量20 km以下的大气风速和气溶胶,风速误差小于2 m s-1,气溶胶的后向散射系数相对误差小于30%.在白天工作时,相同误差下的可探测高度为10 km.模拟分析结果表明,该HSRL雷达有较大的应用前景,对天气和气象研究等有重要意义.     

基于机载观测资料订正雷达比及其垂直分布特征

雷达比是激光雷达反演气溶胶光学特性的重要参数和影响因素。利用北京地区2016年一次清洁过程（12月10日）和两次污染过程（11月15～18日和12月16～19日）的微脉冲激光雷达、机载浊度计和黑碳仪以及多种地基观测设备,综合研究基于飞机观测订正雷达比的方法及其分布特征。清洁过程地面PM_2.5浓度低于40 μg m?3;污染严重时期的PM_2.5均高于150 μg m?3且能见度低于5 km,污染过程1存在高空传输的特征。研究结果表明相较于采用单一的柱平均雷达比,利用本文方法获得的雷达比垂直廓线反演得到的气溶胶消光系数和光学厚度更接近原位跟踪观测,精度均有提升。基于此方法获得的雷达比在污染发展不同时期垂直分布差异较大,主要分布在19～76 sr之间,清洁时期雷达比较小且垂直分布差异不大。污染过程1雷达比随高度波动增加至边界层顶（19～45 sr）;污染过程2严重期边界层内雷达比随高度由70 sr降低到20 sr;边界层以上均呈现小幅波动变化。边界层内雷达比垂直分布与气溶胶来源特别是高空气溶胶传输有密切联系,混有沙尘的区域传输显著提升了所在高度的雷达比值。边界层以上雷达比受少量大粒子或者强吸收性的气溶胶粒子的影响波动变化。边界层内消光系数增大时雷达比呈增加趋势;当相对湿度高于40%,边界层内雷达比随相对湿度增加而增大。     

激光遥测大气消光系数的误差分析

激光遥测大气消光系数分布的主要方法,是假设大气消光系数与大气体后向散射微分截面之比值k为常数,并取激光实测的地面值,然后由解光雷达方程求得。本文将讨论由于大气气溶胶物理属性随高度变化以及大气分子散射的作用,使比值k产生相应变化,并分析由此引起大气消光系数的探测误差。     

霾期间上海低层气溶胶微物理属性与地面相对湿度分析

用2007年1月至2010年11月美国国家航空航天局(NASA)的CALIPSO星载激光雷达资料,通过532 nm总后向散射系数、体积退偏比和色比,分析了上海地区霾期间低层0~2 km高度气溶胶微物理属性与地面相对湿度的关系。结果表明,当地面相对湿度为65%~80%时,后向散射系数在0~0.001 km-1·sr-1范围内出现的频率最大;当地面相对湿度为50%~65%时,后向散射系数大于0.001 km-1·sr-1的频率最大;低层大气中聚集的主要是规则气溶胶粒子,当地面相对湿度为65%~80%时,气溶胶粒子最为规则,其次为80%~95%的地面相对湿度。当地面相对湿度为50%~65%时,不规则气溶胶粒子所占比重较大;霾期间,低层大气中细粒子气溶胶均占主导地位,但随着地面相对湿度的增大,气溶胶粒子粒径逐渐增大。在地面相对湿度为80%~95%时,大颗粒气溶胶相对较多。