利用T-matrix计算非球形粒子散射特性的研究

本文用T-matrix方法计算了非球形气溶胶粒子的光学特性,得到了气溶胶粒子的消光截面、散射截面、吸收截面与气溶胶粒子形状的关系,不同形状气溶胶粒子的有着相同的散射相函数和不同的偏振度,非球形气溶胶粒子的散射相函数对其复折射指数的实部和虚部都不太敏感,而偏振相函数对其实部和虚部都比较敏感.此结论为研究大气辐射传输提供了较好的方法,尤其是偏振度与偏振相函数的提出为用偏振的方法进一步的反演气溶胶的光学参数提供了理论基础.     

用Mie散射理论计算大气气溶胶光学特性

本文在已知厦门海域气溶胶粒子谱分布的基础上,用MIE散射理论计算了该地区气溶胶粒子的光学特性,得到了气溶胶粒子的散射截面、吸收截面、消光截面与波长的关系,散射相函数以及偏振相函数,并用实验对两个相函数加以了验证。此结论为研究大气矢量辐射传输提供了很好的方法;尤其偏振相函数的提出为今后用偏振的方法研究大气提供了理论依据。     

改进的单次散射相函数解析表达式

单次散射相函数对电磁辐射传输模拟过程的准确性和计算效率有重要的影响.基于电磁散射与辐射传输中的基本理论,对单次散射相函数的解析表达式进行了研究,提出了一种新的单次散射相函数解析表达式.比较了单个粒子的Henyey-Greenstein相函数、Henyey-Greenstein*相函数与新的相函数随角度的分布,发现新的散射相函数提高了后向散射峰值,可以更合理地描述单个粒子的散射特性.按三种气溶胶粒子谱分布模式计算了Henyey-Greenstein*相函数和新的相函数对应的数值结果,并与多分散系Mie散射相函数进行对比,发现新的相函数提高了与多分散系Mie散射相函数的符合程度.研究表明,对于大角度(大于90°)后向散射,新的相函数与Mie散射相函数均方根差较小的占73.3%,高于Henyey-Greenstein*相函数的26.7%,证明了新的相函数可以显著提高后向散射峰值.新的相函数对准确模拟辐射传输过程具有重要意义.     

非球形气溶胶粒子散射相函数经验公式

散射相函数是研究电磁波传输特性的重要参数，直接影响电磁波传输方程的简化程度和解的精度。基于电磁散射与辐射传输中的基本理论，对非球形粒子散射相函数的经验公式进行了研究。为了很好的模拟非球形粒子的后向散射峰值，提高辐射传输方程的简化程度和解的精度，提出了一种新的相函数经验公式。分析新的相函数对非球形粒子的适用性，以单个沙尘性气溶胶为例，计算了不同形状粒子的Henyey-Greenstein*相函数和新的相函数随角度的变化，并与T矩阵法的计算结果进行了对比，发现椭球形粒子的长短轴比和有限长圆柱形粒子的径长比大于0.5时，新的相函数在大角度后向散射部分与T矩阵法的吻合程度较高。考虑波长变化，对比了尺寸谱满足对数正态分布的四种气溶胶粒子的Henyey-Greenstein*相函数和新的相函数与T矩阵法的计算结果。研究表明，对于椭球形粒子和有限长圆柱形粒子，在大角度(大于90°)后向散射部分，除了0.694时的椭球形海洋性气溶胶，新的相函数均方根差较小的占100%，证明了新的相函数可以较好的模拟非球形粒子的后向散射特征。新的相函数对准确模拟辐射传输过程具有重要意义。     

基于电荷耦合器件探测气溶胶散射相函数与大气能见度的研究

提出一种测量大气气溶胶散射相函数及能见度的方法,并且设计了以半导体激光器为光源和以电荷耦合器件(CCD)为探测器的实验装置,利用该实验装置测量了15°～45°的散射灰度值角分布。根据激光雷达方程和Henyey-Greenstein散射相函数理论,拟合气溶胶的相函数分布,计算其能见度。将拟合得到的相函数和计算的能见度的结果分别与POM天空辐射计和Belfort能见度得到的观测结果进行对比,结果表明两者一致性较好,证明了利用激光光源和CCD测量气溶胶相函数和能见度的可行性。     

非球形生物气溶胶对偏振光的多次散射特性研究

基于回转椭球模型和有限长圆柱模型,采用T矩阵方法研究了非球形生物气溶胶的单次散射特性,计算了鼠疫耶尔森氏杆菌、土拉热杆菌二种生物气溶胶对氦氖激光的单次相矩阵、单次散射反照率以及不对称因子。根据矢量辐射传输理论,研究了激光在生物气溶胶中传输的偏振散射特性,基于累加-倍加法(adding-doubling method)求解矢量辐射传输方程,并计算了非球形生物气溶胶对激光多次散射的斯托克斯参量。计算结果表明,生物气溶胶的尺寸和形状对光的极化更为敏感,因此在利用激光进行生物气溶胶微观特性探测和反演时,利用激光的偏振散射特性为非常有效的方法。     

多次散射辐射传输计算中的相函数处理

在实际大气中,分子和粒子不仅遭受单次散射,还遭受多次散射。本文首先计算了多次散射在总散射量中的贡献。结果表明当光学厚度大于0.1时,多次散射在总散射量中的贡献将超过10%,因此必须考虑多次散射的作用。在多次散射辐射传输计算时,一般需把散射相函数展开为勒让德(Legendre)函数的多项式。有些介质如云或气溶胶粒子,散射相函数前向非常尖锐。展开的Legendre多项式需数百项甚至上千项才能收敛,而计算时间与展开项数的3次方成正比。本文介绍了在辐射传输计算时对尖锐相函数的δ-M和δ-fit处理方法,比较了两种方法的计算结果。当计算用的流数相同时,δ-fit方法的计算结果比δ-M方法的计算结果要精确得多;当计算结果精确度相同时,δ-fit方法的流数比δ-M方法的流数少得多,运算速度也快很多。δ-fit方法是目前处理散射相函数的理想方法,可以大大提高计算的精度以及缩减运算时间。     

风驱粗糙海面背景下海雾对可见和红外光多次散射特性研究

通过测量空间辐射反演海雾气溶胶的微观特性是一种重要的遥感方法,但是海雾对辐射的反射会受到海面背景的影响。该工作研究了海雾与海面的耦合多次散射。利用Mie理论研究了海雾气溶胶的单次散射特性,利用辐射传输理论研究太阳光在无下垫面海雾中传输多次散射特性;在基尔霍夫近似下研究了风驱粗糙海面的散射特性,并考虑了海面的遮蔽效应,得到了风驱海面随风速的反射函数。根据累加法研究了海雾和海面的耦合散射,计算结果表明海雾在有粗糙海面作为背景的情况下,其整体反射有较大增强。     

包容性非球形气溶胶粒子的散射特性及其对激光传输的影响研究

针对局部地域大气气溶胶的生成、演化的特点,在分析典型对称结构的包容性非球形气溶胶粒子的形态学特征基础上,利用T矩阵法进行了椭球形粒子含非球形粒子的包容性粒子光散射特性计算和分析.以1 064nm波长激光传输为例分别计算了两种典型分布状况的包容性非球形粒子的散射相函数等散射特性,并针对稠密气溶胶大气路径上激光传输的多散射现象,利用蒙特卡罗方法进行了典型大气状况下包容性粒子多次散射对激光信号大气传输中脉冲展宽的应用分析,在传输3km和5km距离时多次散射效应造成激光脉冲峰值半宽分别为3μs和4μs左右.指出气溶胶的基本散射特性与粒子的形态学特征、尺度因子、化学成分及颗粒密度等有密切的关系,其散射特性对激光信号的大气传输具有明显的影响.     

两种按比例混合颗粒系的多次散射模拟

一种颗粒与其他种类的颗粒混合后, 会使其散射特性发生变化, 本文研究了水云中混有黑炭气溶胶后的散射特性变化. 根据Mie理论计算了水云和黑炭气溶胶散射相函数、单次散射反照率和不对称因子. 给出了混合颗粒系的蒙特卡罗模拟方法, 给出了颗粒碰撞类型抽样、自由程抽样和根据Mie相函数进行散射方向抽样的方法. 计算了光垂直入射时, 水云和黑炭气溶胶混合颗粒系的反射光强随观测角的变化, 并计算了平面反照率随入射角的变化, 讨论了黑炭气溶胶的有效半径、混合比例对整个混合颗粒系散射特性的影响. 计算结果表明, 水云中混合黑炭会加强其吸收, 且黑炭的比例和尺寸不同其散射特性差异较大.     

On the finite volume method and the discrete ordinates method regarding radiative heat transfer in acute forward anisotropic scattering media

The ability of the finite volume method (FVM) and the discrete ordinates method (DOM) to model radiative heat transfer in acute forward anisotropic scattering media has been investigated. The test case involves a purely scattering medium in a cubic enclosure, irradiated by one boundary with diffuse emission. Four phase functions have been considered: three of the Henyey-Greenstein type with respective asymmetry factors of 0.2, 0.8 and 0.93, and a Mie phase function with a strong forward scattering peak (computed for a size parameter of 245 and corresponding to an asymmetry factor of 0.93). Results obtained with the FVM are in good agreement with Monte Carlo reference solutions, whatever the level of acute anisotropic scattering (for asymmetry factors up to 0.93). The DOM combined with the renormalization procedures of the phase function proposed by Kim and Lee (Effect of anisotropic scattering on radiative heat transfer in two-dimensional rectangular enclosures. Int J Heat Mass Transfer 1988;31:1711-21. [1]) and Wiscombe (On initialization error and flux conservation in the doubling method. JQSRT 1976;18:637-58. [2]) provides accurate results only for the smallest asymmetry factor. As the asymmetry factor increases, the renormalization procedures induce strong modifications in the values of the discretized phase function resulting in an underestimation of the effective attenuation by scattering. This error has been found to increase with optical thickness. In fact, when using the DOM, results would be more accurate combining this method with a Delta-Eddington approximation of the phase function, instead of using the actual phase function which is altered too much by renormalization.     

A direct solution of the radiative transfer equation: Application to Rayleigh and Mie atmospheres

A direct method is given for the solution of the spherical harmonics approximation to the equation of radiative transfer in plane-parallel atmospheres. Although the method is formulated theoretically for non-homogeneous atmospheres with an arbitrary phase function, at present it has only been implemented for homogeneous atmospheres. Test computations performed for Rayleigh and Mie scattering phase functions show that the direct method is unconditionally stable and solves efficiently problems both for optically thin and very thick atmospheres. Timing comparisons with the method of Chandrasekhar for Rayleigh atmospheres and with an integral-equation iterative method for Mie atmospheres are quite favorable to the proposed method.     

NEW APPROXIMATE PHASE FUNCTIONS FOR SCATTERING OF UNPOLARIZED LIGHT BY DIELECTRIC PARTICLES

We present two new phase functions, one for particles small compared to the wavelength of the scattering radiation and the other for particles large compared to the wavelength of the scattering radiation. These phase functions have been validated for the case of Mie scatterers. For small particles, the results of the new phase function are found to be identical with the Mie results. For large particles, comparison with the Mie results show that the phase function presented here is an extremely good approximation to the Mie phase function. We believe that these phase functions can be expediently used in problems relating to solutions of the radiative transfer equations.     

Transmission of radiation through an aerosol medium

The properties of radiation through an aerosol medium have been achieved. This has been done by employing Mie scattering theory to calculate the radiation transfer scattering parameters in the form of extinction, absorption and scattering efficiencies. The equation of radiative transfer for the heat flux through a plane parallel atmosphere of aerosol has been solved. The aerosol size distributions are found in practical systems. Average efficiencies over size distribution for spherical particles of complex refractive index are calculated. Therefore, the radiative properties of stratospheric aerosols have been done. The obtained results found to be in a good agreement with the previous work.     

Numerical results for the thermal scattering functions

A recent formulation in radiative transfer defined the thermal scattering functions that characterize radiative transfer from a general, plane-parallel, finite medium driven solely by an internal distribution of thermal sources. Exiting diffuse intensities are expressed as space convolutions of the thermal scattering functions with any thermal source distribution. A parametric study is presented to obtain the basic structure of these scattering functions. The independent variables of these azimuthally independent functions are the direction consine μ and source location t, while the parameters are the single scattering albedo ω, total optical depth t₀, and the asymmetry factor g in the Henyey-Greenstein phase function. The basic functional trends are discussed using various parametric plots, and selected tabular results are given to allow numerical checks. The computational method is invariant imbedding. As a particular application, these functions are used in the following companion paper to obtain exiting intensities from inhomogeneous and nonisothermal media.     

Radiative transfer in finite participating atmospheric aerosol media

The properties of radiation transfer through a plane-parallel atmospheric aerosol medium has been studied. It has been done by employing Mie theory to calculate the radiation transfer scattering parameters of the medium in the form of extinction, scattering, and absorption efficiencies. Then, the equation of radiative transfer through a plane-parallel atmosphere of aerosol has been solved for partial heat fluxes using two different analytical techniques, namely, the Variational Pomraning -Eddington approximation and Galerkin technique. Average efficiencies over log-normal and modified gamma size distributions are calculated. Therefore, the radiative properties of Carbon, Anthracite, Bituminous, Lignite, and Fly ash have been calculated. The obtained numerical results show very good agreement with each other in addition to the previous published work.     

Radiative transfer in plant leaves

A two-layer model of light scattering and absorption in plant phytoelements is considered, which takes into account absorption of light by pigments and water and light scattering by particles of two types: chloroplasts and air cavities. An elementary light scattering event is described using the Mie theory. Multiple light scattering in a leaf is described within the framework of the theory of radiative transfer. The equation of radiative transfer with a strongly anisotropic phase function is solved using the method of addition of layers and the method of reduction to a medium with effective parameters depending on the propagation direction of light. The spectral dependences of reflection and transmission coefficients are calculated in the visible range as functions of the leaf structure.     

Quadrature effects on the accuracy of flux calculations in realistic atmospheres

We have investigated the accuracy of five different quadrature methods—equal steps in θ, equal steps in cos θ, Gaussian, double Gaussian and Gauss-Lobatto—on the accuracy of fluxes in realistic aerosol atmospheres, using the Gauss-Seidel method. In addition, a range of Gaussian quadrature stream numbers from two to 32 were compared. The atmospheric models considered are those recently presented by Lenoble, with the exception that we have used Henyey-Greenstein phase functions in place of Mie. Our results should be easily reproduceable by any other workers interested in similar realistic atmospheres. A table of Gauss-Lobatto weights and points is provided as an appendix.