Aerosol concentration measurement with multiple light scattering system and laser aerosol spectrometer

The performance of two different optical concentration-measuring techniques was investigated over a concentration range starting with about 10² cm⁻³ and extending over more than four decades. Both instruments are capable of real-time counting, however due to their particular design-single particle counter and ensemble particle-measuring system—they operate in overlapping, but different concentration ranges. The upper, coincidence-free counting limit for the single particle counter used in this study was established to be in the order of 10⁴ cm⁻³. The ensemble technique was found to be functional and stable for concentrations of about 10³ cm⁻³ and limited by the onset of multiple scattering at concentrations nearby 2×10⁶ cm⁻³. Within the determined boundaries, both techniques proved to provide reliable aerosol concentration data.     

半无限空间中稳态P波在衬砌周围的散射

本文采用多级坐标和复变函数方法针对半无限空间中P波在一衬砌周围的散射问题提出了一种近似求解分析方法。具体做法是利用一个半径很大的圆来逼近半空间的直边界，将待解问题转化为全空间中P波在一圆孔和一衬砌周围的共同散射问题。我们预先写出问题波函数的一般形式解，利用边界条件并借助复数傅立叶级数展开把问题化为求解波函数中未知系数的无穷线性代数方程组，进而讨论了衬砌内外边界处动应力集中系数针对不同条件组合的分布和变化情况。算例结果表明：本文方法对于研究与P波（SV波）有关的散射问题是可行的。     

高分辨率SAR数据在土地利用监测中的应用研究

随着高分辨率SAR卫星的陆续发射,其全天时、全天候、高效、准确获取地表数据的能力日益展现出明显的优势。在当前土地资源管理不断要求加快监测周期、扩大应用领域、提高监测成效的紧迫形式下,对SAR数据的深入研究与工程化应用已成为当务之急。本文以土地利用的监测实践为依据,系统分析了SAR传感器的数据基础、影像特征和基于实际应用而需要继续拓展的研究领域。     

Scattering process of internal waves propagating over a subcritical strait slope onto a shelf region

The scattering process, which means the redistribution of energy fluy in modenumber space, is analyzed for internal waves propagating from the abyssal ocean onto a subcritical strait slope and then a shelf region. In light of Wunsch＇s work, the waves are analytically expressed as superimposition of eigensolutions distribution of energy flux in the shelf region： one is the ratio of water depth in and the other is the ratio of the slope of the internal tide rays to the topographic energy flux distribution： the energy flux is focused around one modenumber or case, the range of modenumbers where energy flux is distributed is narrow. Two parameters have evident effects on the the shelf region to that in the abyssal ocean slope. Generally, there are two patterns of focused around two modenumbers. In any case, the range of modenumbers where energy flux is distributed is narrow.     

宽带太阳漫射辐射法反演辐射加权平均的气溶胶一次散射反照率研究

引入了一个辐射加权平均的宽带气溶胶一次散射反照率(SSA)的定义,提出了一个从宽带的漫射信息反演该SSA的方法.数值模拟结果表明,在Junge气溶胶谱分布情形下,对气溶胶光学厚度、(A)ngstr(o)m 指数与气溶胶虚部的通常变化范围,应用该SSA所计算的2160组宽带太阳辐射反射率、漫射透过率、总透过率和吸收率的相对标准误差都在1.107%以内; 绝对标准差在0.00287以内.对非Junge的大陆性和都市工业污染气溶胶模式(由水溶性、沙尘和碳粒子组成),在72组反射率、漫射透过率、总透过率和吸收率计算中,相对标准差都在2.047%以内,绝对标准差在0.0075以内.在Junge气溶胶假设下,作者提出了一个综合应用宽带太阳直射和漫射信息同时反演气溶胶光学厚度与辐射加权平均SSA的方法,并通过模拟反演分析了SSA反演的3个主要误差因子.从反演结果可以看出:(1)如果(A)ngstr(o)m指数误差在±0.2以内,对0.55 μm气溶胶光学厚度大于0.312大陆性气溶胶,SSA误差在±0.0418以内;(2)波长无关的宽带地表反照率适用于SSA反演;(3)气溶胶光学厚度越大,辐射资料误差所引起的SSA解误差越小.当辐射误差在±2%以内以及0.55 μm气溶胶光学厚度大于0.312时,SSA解的误差在±0.0149以内; 辐射误差在±4%以内时,SSA解的误差在±0.0317以内.     

SCATTERING OF SH WAVES IN MULTILAYERED MEDIA WITH IRREGULAR INTERFACES

An indirect boundary integral equation method is used to investigate the scattering of elastic SH waves in a multilayered media with irregular interfaces. An extensive parametric error analysis is performed in order to assess the convergence of the proposed method. Subsequently, the numerical results for both steady-state and transient surface response are given for one- and two-layer models.     

云南地区尾波衰减特征研究

利用22个云南区域数字地震台网近震资料,研究了云南地区地震尾波在1~20 Hz内6个频段的衰减特征. 首先使用单次散射模型计算各台站的尾波衰减系数; 然后利用随机各向同性散射介质均匀分布的多次散射模型, 使用多流逝时间窗分析方法,计算各台站的平均自由程Le和地震反照率B0得到总衰减系数,并从中分离散射衰减系数 、吸收衰减系数 . 云南地区基本衰减特征是Le主要集中在10~30 km内, 在2~6 Hz之间有最大值;B0在1~2 Hz大致为0.5左右,在其它频段则小于0.5. 也即在1~2 Hz, 与 数值相当, 其它频段 逐渐大于,成为尾波总衰减的主要原因. 除1~2 Hz外,其它频段 值接近. 云南地区的Q-1尤其是, 在空间上有较大的差异,滇东略微高于滇西北, 滇西北高于滇西南地区. 与全球其它区域的研究结果比较发现,云南地区 略低于平均水平, 略高于平均水平, 而则处于其它区域结果的中间.      

基于权重散射特征的FSVM算法在极化SAR影像分类中的应用

提出了一种基于权重散射特征的模糊支持向量机的极化SAR数据监督分类方法。首先,对极化SAR数据进行H/SPAN/A/α散射特征提取;其次,根据样本可分离度设置最佳散射特征权重参数C,得到最优分类输入数据(H/6SPAN/A/α);最后,利用FSVM算法对数据进行监督分类。实验结果证明,所提出的FSVMH/6SPAN/A/α分类结果优于FSVM-H/SPAN/A/α。