Application of wavelet transformation to determine wavelengths and phase velocities of gravity waves observed by lidar measurements

During the last two decades, important advances have been made in the investigation of gravity waves. However, more efforts are needed to study certain aspects of gravity waves. In the real atmosphere, gravity waves occur with different properties at different altitudes and, most often, simultaneously. In this case, when there is more than one dominant wave, the determination of gravity wave characteristics, such as the vertical wavelength and the phase velocity, is difficult. The interpretation of temperature perturbation plots versus the altitude and time as well as the application of the Fourier spectral analysis can produce errors.Exact knowledge of the wave characteristics is important both for determination of other characteristics, for example, the horizontal wave components, and for study of wave climatology. The wavelet analysis of vertical temperature profiles allows one to examine the wave's location in space. Up to now, gravity waves have been studied mainly by continuous wavelet transformation to determine dominant waves. We apply wavelet analysis to a time series of temperature profiles, observed by the ALOMAR ozone lidar at Andoya, Norway, and by the U. Bonn lidar system at ESRANGE, Sweden, both for determination of the dominant waves and for specifying the vertical wavelengths and the vertical component of the phase velocities. For this purpose, the wavelet amplitude spectra and the wavelet phase spectra are filtered and Hovmöller diagrams for dominant wavelengths are constructed. The advantage of this type of diagrams is that they give clear evidence for the localization of the dominant waves in space and time and for the development of their phase fronts.     

Sagnac型干涉成像光谱仪研制及干涉成像光谱实验

空间调制干涉成像（傅里叶变换）光谱仪（SMIFTS）是一种有重要的研究和应用价值的空间遥感器。依据应用要求和信噪比方程设计并研制了Sagnac型干涉成像（傅里叶变换）光谱仪原理样机。用该样机进行了实验室干涉成像光谱实验，获取了较为清晰的可见光、近红外谱段干涉成像图像。对其中各像元的干涉图进行数据处理和计算，提取出了光谱特征明显的成像光谱图像和与被测图样光谱特征相符的像元光谱。由实测光谱辐射亮度曲线反演计算出的光谱反射曲线与标准仪器测得的图样光谱反射特性曲线进行了比较，结果表明二者对应相似。     

Linear drift and periodic variations observed in long time series of polar motion

 Two long time series were analysed: the C01 series of the International Earth Rotation Service and the pole series obtained by re-analysis of the classical astronomical observations using the HIPPARCOS reference frame. The linear drift of the pole was determined to be 3.31 ± 0.05 milliarcseconds/year towards 76.1 ± 0.80° west longitude. For the least-squares fit the a priori correlations between simultaneous pole coordinates x _p , y _p were taken into account, and the weighting function was calculated by estimating empirical variance components. The decadal variations of the pole path were investigated by Fourier and wavelet analysis. Using sliding windows, the periods and amplitudes of the Chandler wobble and annual wobble were determined. Typical periods in the variable Chandler wobble and annual wobble parameters were obtained from wavelet analyses. Received: 21 January 2000 / Accepted: 28 August 2000     

Bivariate Wavelet Analysis of Asia Monsoon and ENSO

This paper employs some recently developed bivariate wavelet analysis techniques to study the correlation between Asia monsoon and E1 Nino southern oscillation (ENSO).Various energy spectral densities are defined for waveiel transforms,analogous to those used in conventional Fourier analysis.Some comparisons are made by applying both wavelet and Fourier spectral methods (o the data.The wavelet analysis shows evidence of some relationship between Asia monsoon and ENSO,which the Fourier analysis resolves poorly.Correlation on several time scales,ranging from 2-4 years,11 years,and 22 years,become apparent with the wavelet cross-spectrum.Finally,the warelet cross-transform provides time localization of the distinctive features within the data record.     

Wavelet analysis of the standard map: Structure and scaling

Wavelet analysis is applied to distributions of points generated by iterating the standard map. The initial condition is chosen so that the points fill the largest chaotic region. When the standard map parameterk=1.3, the distribution of points contains many voids corresponding to islands in the chaotic region. The wavelet transform is dominated by contributions from these islands. Fork=10 the chaos fills phase space and no structure is apparent; the wavelet transform reveals statistical fluctuations in the distribution of points.     

甘肃河东年降水量的周期变化

利用甘肃河东建站最早的９个代表站的年降水量资料,采用谐波分析和小波分析方法,计算了年降水量序列各种时间尺度的周期,以及１０ａ和２０ａ相空间时间序列的２～１０ａ周期试验;陇南无明显周期;陇南无明显周期。相空间序列的周期试验显示,甘肃中部的准３ａ周期主要反映在５０～７０年代,近２０～３０ａ则以８～９ａ周期较为明显;陇东近１０ａ主要表现为准２ａ振荡。进一步的分析表明,甘肃中部年降水量的准３ａ周期振荡目处     

Wavelet filter analysis of atmospheric pressure effects in the long-period seismic mode band

The importance of the reduction of atmospheric pressure effects becomes very clear when investigating seismic normal-mode spectra below 1.5 mHz. The usual simple correction method consists in subtracting a term converted from local atmospheric pressure (pressure multiplied by a frequency-independent admittance) from the gravity record in time domain. Thus, estimating an efficient admittance is the key for an improved correction. Band-pass filters derived from dyadic orthogonal wavelet transform, having narrow pass-bands with good frequency response but without Gibbs phenomenon and causing no phase lag, are very helpful to estimate an efficient admittance, which is both time and frequency-dependent. Processing of high quality superconducting gravimeter (SG) records for the great Sumatra earthquake (Mw = 9.3, Dec 26, 2004) with wavelet filters reveal the three very well resolved splitting singlets of overtone ₂S₁ with a single gravity record after correction with time-dependent and frequency-dependent admittances. We also observe all coupled toroidal modes below 1.5 mHz, except ₀T₅, ₀T₇ and ₁T₁, with good signal-to-noise ratio (SNR); moreover, toroidal modes ₁T₂ and ₁T₃ are for the first time unambiguously revealed in vertical components.     

大欧拉角的空间直角坐标转换方法探讨

首先指出建立使用大旋转角的空间直角坐标转换模型的必要性，接着推导出以罗德里格矩阵为旋转矩阵的大欧拉角空间直角坐标转换的数学模型，然后以布尔沙模型为例、以Visual C#．NET2003编制了计算软件。该算法具有性能稳定、计算收敛速度快、精确度高、转换参数初始值可以为任意值的优点。     

夏季青藏高原东部降水变化与副热带高压带活动的研究

用青藏高原上常规的台站资料和NCEP/NCAR的再分析资料,分析和研究了1993/1994年夏季5—8月青藏高原东部降水变化与西太平洋副热带高压南北移动的关系。结果表明:1993年夏季的副热带高压异常偏南,5—8月副高脊线北移过程中,还伴有准双周南北移动过程;而1994年夏季的副热带高压异常偏北,5—8月副高脊线北移过程中,则主要伴有30—60d的低频变化的南北移动过程。同时,在分析青藏高原上的天气变化特征时,发现这两年高原东部的降水变化特征也有明显的不同,夏季高原上降水的活跃和中断与副高脊线南北移动变化有类似的特征。因此,夏季西太平洋副热带高压南北移动可能与高原东部降水的中断/活跃有一定的关系。     

用小波包识别地震和矿震

在能检测到天然地震和矿震的区域，这两类地震的快速识别无论对于区域台网和矿区台网都具有现实意义。这两类震动都是非稳态信号，用传统的Fourier变换不能提取出信号的特征信息，小波包分析方法却能很好提取出信号的特征信息。本文提供了一种基于非参数识别算法，即把信号变换到频域，然后再用奇异值分解作为统计工具，提取出信号的特征信息，作为识别天然地震和矿震的识别因子。以辽宁抚顺2001年1月1日到2003年6月30的18个矿震和16个天然地震，以及北京门头沟2001年1月1日到2002年12月31日的15个矿震和14个天然地震为样本，提取出识别因子。最后，用其它的天然地震和矿震资料检验了识别因子的识别率。