一种快速预处理算法及其在浅水波方程中的应用

为了提高数值求解大气方程的计算速度,我们研究了稀疏逼近逆预处理方法及其在数值求解浅水波方程中的应用,这是一种求解大型线性方程组的快速算法,该算法的核心内容是稀疏逼近逆非零元模式的选取。首先导出了一种稀疏逼近逆的非零元模式及其确定方法,然后以浅水方程的差分格式为例,借助于GMRES迭代算法,对这种预处理快速算法应用前后的计算速度进行了比较,发现该快速算法能大幅度提高运算速度。另外,该预处理快速算法简单、易于并行,是一种值得在大气方程中推广应用的方法。     

单多普勒天气雷达反演降水粒子垂直速度Ⅰ：算法分析

针对多普勒雷达风场反演的体积速度处理(Volume Velocity Processing,VVP)方法中系数矩阵的病态问题,从数学上进行了分析和论证,对反演的误差进行了敏感性分析,并对垂直速度的求解方程作了改进。系数矩阵的条件数将随着待反演参量的不同而差别很大,通常的处理方法是舍弃量级较小的参量,通过对误差范数的分析,证明这种处理尽管存在模型误差,但能够降低求解难度和结果误差。对系数矩阵病态原因的分析发现,系数矩阵矢量的线性相关造成了矩阵奇异,当合并或舍弃线性相关项时,待反演参量会受模型误差的影响,并且这种模型误差的大小并不仅与待反演参量的量级有关,而且随着位置的不同而改变,但是部分待参量仍然可以保持准确值。在对VVP算法误差分析的基础上,分析并验证了舍弃部分参量时的反演误差,改进的算法为准确反演降水粒子的垂直速度提供了理论基础。     

预处理JFNK方法及其在大气方程中的初步应用

为提高数值求解大气方程的速度, 研究了预处理JFNK（Jacobian-Free Newton-Krylov）方法及其在大气方程中的应用。这是一种非线性外循环Newton迭代与线性内循环Krylobv迭代相结合的快速算法,其优点是进行外循环Newton迭代时不要求Jacobian矩阵的形成和存储;它的有效性取决于内循环中线性系统的预处理。首先介绍了JFNK算法,然后以浅水波方程为例,描述了非线性残值的形成、预处理矩阵的构造及其在JFNK算法中的应用。试验结果表明：对内循环线性系统进行适当的预处理,能大幅度提高JFNK算法的运算速度。因而,JFNK是一种值得在大气方程中推广应用的方法。     

准地转三层模式天气值预报方法的试验研究

作者首先根据温度局地变化的分析说明三层模式常常是必要的。然后作者用较普遍的准地转三层模式,试算了大范围高空温压场的倾向。所计算的是乌拉尔阻塞流型发展关键时刻的例子。作者的计算结果说明这种阻塞流型形成的预报必须使用三层模式,主要是因为与阻塞流型中高、低压切断有关的负变高,只有使用三层模式才报得好。根据三层模式的公式,作者并求出大范围天气系统发展的规则:即冷(暖)平流向上增强(减弱),那末等压面的位势高度升高(低槽减弱、高压脊加强);冷(暖)平流向上减弱(增强),那末等压面的位势高度降低(低槽加强、高压脊减弱)。也就是说,—舵所接受的暖平流使脊加强,冷平流使槽加深这规则是有条件的。作者还指出风场的逼近的需要和困难。特则是现有高空风观测和电码难以直接用来计算辐散。在计算方面作者给出了一个对 Poisson 方程适用的成群放松的格式。     

Low-level wind response to mesoscale pressure systems

Observations are presented which show a strong correlation between low-level wind behaviour (e.g., rotation near the surface) and the passage of mesoscale pressure systems. The latter are associated with frontal transition zones, are dominated by a pressure-jump line and a mesoscale high pressure area, and produce locally large horizontal pressure gradients. The wind observations are simulated by specifying a time sequence of perturbation pressure gradient and subsequently solving the vertically-integrated momentum equations with appropriate initial conditions. Very good agreement is found between observed and calculated winds; in particular, (i) a 360 ° rotation in wind on passage of the mesoscale high; (ii) wind-shift lines produced dynamically by the pressure-jump line; (iii) rapid linear increase in wind speed on passage of the pressure jump.     

有限区域风速场求解流函数和速度势场的有效方案

流函数和速度势是表示风场的一种变量, 在数值天气预报模式和分析、同化方案中经常使用, 通常可以用风速分量场求解Poisson方程得到。对于有限区域系统, 往往采用差分方法, 但由于存在边界问题, 用计算所得到的流函数和速度势场重建风速场, 在边界附近经常出现明显的偏差。基于差分方法、利用有限区域风速场求解流函数和速度势场的基本方法和特点的分析, 在Arakawa A网格分布的有限区域, 设计了一种用差分方法求解流函数和速度势场的有效方案。在该有效方案中, 通过将有限区域向外扩展二圈, 风速场线性外推, 改进计算边界风速值和边界定解条件的效果; 尽可能使用协调、一致的差分格式, 提高求解精度; 最后利用一种增量订正迭代方法, 迭代2~3次就可以获得令人满意的结果。实例试验的对比、检验显示, 用该方案计算求得的流函数和速度势场重建风速场, 具有非常高的精度。     

Improved Estimation of Errors due to Antenna Geometry in RASS Based on a Radar Wind Profiler

Summary A rigorous theoretical consideration of the parameters of a received signal for a RASS based on a radar wind profiler and operating in the “Bragg”-operation mode is performed. The calculation approach is based on solving the equations analytically for diffractive wave beams and takes into account diffraction effects in the Fresnel approxim ation for both radio and acoustic waves generated by antennas with finite sizes. It is shown that the errors due to the displacement of an acoustic source and horizontal wind are negligible for RASS based on a monostatic radar not only in the far-field but also in the transition zone between the Fresnel diffraction and far-field ranges at the lowest RASS operational heights. Received August 24, 1998 Revised February 25, 1999     

Experience of using a dynamic stochastic approach for solving the problem of very short-range forecasting of state parameters of the atmospheric boundary layer

Discussed are the results of applying a dynamic stochastic method based on the use of the two-dimensional model and the Kalman filtering algorithm for solving the problem of the very short-range (from 0.5 to 6 hours) fore cast of air temperature and orthogonal components of the wind speed in the atmospheric boundary layer realized using the data of radio metric, sodar, and in creased-frequency radiosonde measurements. It is demonstrated that the pro posed technique and the appropriate algorithm give a rather high ac cu racy of very short-range fore casting of temperature and wind within the lead time range under consideration.     

Nelder-Mead单纯形算法改进及在气象上的应用

Nelder-Mead Simplex (NMS)算法是一种查找多元函数局地最小值的无微分算法,在现代科学计算中得到广泛应用,该文提出了一种对NMS算法的改进方法.改进后,大大简化了其计算过程,提高了该算法的收敛速度.利用改进后的算法对陆面过程参数进行了拟合计算,结果表明:改进的NMS算法对非线性公式具有非常高的拟合精...     

变分方法反演双多普勒雷达低层二维风场

提出用变分方法和正则化方法相结合反演双多普勒雷达资料的低层二维风场的方法,即在目标泛函中加入正则化项（先验信息项）且合理选择正则化参数。数值试验结果表明：正则化项的引入能有效滤除反演风场中因奇点和观测资料引起的误差,提高反演精度,得到最优的全局反演风场。     

风云二号气象卫星图像定位和卫星风精度的改善中解决问题的途径

本文以改善风云二号气象卫星图像定位和卫星遥感风精度的工作为例,简述工程技术任务中解决问题的途径.     

An Improved Velocity Volume Processing Method

Velocity volume processing （VVP） retrieval of single Doppler radar is an effective method which can be used to obtain many wind parameters. However, due to the problem of an ill-conditioned matrix arising from the coefficients of equations not being easily resolved, the VVP method has not been applied adequately and effectively in operation. In this paper, an improved scheme, SVVP （step velocity volume processing）, based on the original method, is proposed. The improved algorithm retrieves each group of components of the wind field through a stepwise procedure, which overcomes the problem of an ill-conditioned matrix, which currently limits the application of the VVP method. Variables in a six-parameter model can be retrieved even if the analysis volume is very small. In addition, the source and order of errors which exist in the traditional method are analyzed. The improved method is applied to real cases, which show that it is robust and has the capability to obtain the wind field structure of the local convective system. It is very helpful for studying severe storms.     

Flow and Transport in the Obstacle Layer: First Results of the Micro-Scale Model MITRAS

A new atmospheric micro-scale chemistry, transport and stream model, MITRAS, has been developed within the Tropospheric Research Program (TFS). MITRAS is an obstacle resolving model solving simultaneously the governing equations for chemistry, pollutant transport including soot as well as for the flow and temperature field. The chemical reaction module is based on a simplified RACM mechanism. MITRAS has been tested against quality ensured wind tunnel data. The model results on soot and ozone concentrations as well as on the spatial representativeness of measurement sites underline the necessity for high-resolution flow simulations for case studies on close-to-source chemistry and pollutant transport within the urban canopy layer.     

夏季东亚副热带西风急流的季节内变化及其与我国东部降水的关系

利用ERA-interim再分析资料和中国降水观测数据,分析了夏季东亚副热带西风急流的季节内变化以及对我国降水的可能影响,结果表明:亚洲地区夏季200 hPa纬向风在空间上主要表现为南北振荡和东西振荡特征,在时间上具有10～40 d的周期;在低频尺度(10～40 d)上,纬向风异常由西北先向东传播,到达东亚地区后再向南传播;伴随低频纬向风季节内演变,高原以北的急流中心向东和东南移动,急流轴也呈现南北振荡的特征;降水异常对200 hPa风场低频振荡有显著响应,在东亚地区,低频纬向风与低频降水南移,降水在黄淮地区出现,并逐渐移至长江及其以南地区;急流附近的低频西(东)风异常在其南侧形成负(正)切变涡度,对应我国东部地区位势高度的升高(降低),使得南亚高压东伸(西退),从而使我国东部雨带的位置发生显著变化。     

风廓线雷达产品数据置信度研究

随着风廓线雷达技术的发展,高空风探测参量越来越多,数据精度不断提高,探测能力得到了极大提升。评估风廓线雷达数据置信度是风廓线雷达应用中需要解决的重要问题。本文基于径向数据和风场合成两个阶段,在风廓线雷达数据反演过程中形成数据置信度算法,并引入噪声电平。同时,利用南京同站址风廓线雷达和探空1 a的资料进行匹配比对,对置信度算法性能进行评估,结果表明该置信度算法可行。将置信度算法植入风廓线雷达数据处理软件中,能实时输出含置信度的风廓线雷达产品数据,有利于预报人员合理使用置信度较高的风廓线雷达产品数据。对于置信度较差的产品数据进行分析,可有助于及时发现雷达的潜在故障。     

IMPLICIT NONLINEAR NORMAL MODE INITIALIZATION: A MULTIGRID APPROACH

The implicit nonlinear normal mode initialization (INMI) is applied to a tropical limited area shallow water model in spherical coordinates.The boundary condition for the INMI scheme is based on the boundary formulation of the model.The INMI scheme is found to be very efficient in suppressing spurious gravity wave oscillation and providing a well balanced initial data set for the model.The INMI scheme involves solving a number of elliptic type equations with varying complexity.and hence an efficient numerical technique is required for solving such equations.In order to make INMI computationally more attractive,we are employing the multigrid method for solving all the elliptic type equations in the INMI scheme.The numerical procedures for the development of such multigrid solvers are briefly described.     

IMPLICIT NONLINEAR NORMAL MODE INITIALIZATION:A MULTIGRID APPROACH

The implicit nonlinear normal mode initialization (INMI) is applied to a tropical limited area shallow water modelin spherical coordinates.The boundary condition for the INMI scheme is based on the boundary formulation of themodel.The INMI scheme is found to be very efficient in suppressing spurious gravity wave oscillation and providing awell balanced initial data set for the model.The INMI scheme involves solving a number of elliptic type equations withvarying complexity.and hence an efficient numerical technique is required for solving such equations.In order to makeINMI computationally more attractive,we are employing the multigrid method for solving all the elliptic type equationsin the INMI scheme.The numerical procedures for the development of such multigrid solvers are briefly described.     

斜压气流的中尺度稳定性 Ⅰ.对称不稳定

本文是“斜压气流的中尺度稳定性”的第一部份,讨论斜压基流对平行型中尺度扰动的稳定性,寻找弱稳定层结大气中出现带状中尺度扰动的可能性及其对深厚对流云系的启动和组织作用。 用非弹性近似(anelastic assumption),导出适于用广义矩阵法及打靶法解特征值问题的理想大气扰动方程,并在线性风,双曲正切型风及飑线发生前的实测风分布下,计算出中尺度不稳定谱及特征扰动结构,通过能量分析,说明斜压性在产生这种非地转惯性对流中的作用。     

A note on smoke concentration prediction in Athens,Greece

Summary In this paper an attempt for predicting smoke concentration levels in Athens for spring and autumn is presented. The regression models are based on equations relating wind speed and smoke values. It is found that both of them become more effective if corrections related to wind speed categories are applied to the equations. In this way good prediction is achieved for the next 24 hours.With 4 Figures     

Gust Front Statistical Characteristics and Automatic Identification Algorithm for CINRAD

Gust front is a kind of meso-and micro-scale weather phenomenon that often causes serious ground wind and wind shear. This paper presents an automatic gust front identification algorithm. Totally 879 radar volume-scan samples selected from 21 gust front weather processes that occurred in China between 2009 and 2012 are examined and analyzed. Gust front echo statistical features in reflectivity, velocity, and spectrum width fields are obtained. Based on these features, an algorithm is designed to recognize gust fronts and generate output products and quantitative indices. Then, 315 samples are used to verify the algorithm and 3 typical cases are analyzed. Major conclusions include： 1） for narrow band echoes intensity is between 5 and 30 dBZ, widths are between 2 and 10 km, maximum heights are less than 4 km （89.33%are lower than 3 km）, and the lengths are between 50 and 200 km. The narrow-band echo is higher than its surrounding echo. 2） Gust fronts present a convergence line or a wind shear in the velocity field;the frontal wind speed gradually decreases when the distance increases radially outward. Spectral widths of gust fronts are large, with 87.09% exceeding 4 m s-1 . 3） Using 315 gust front volume-scan samples to test the algorithm reveals that the algorithm is highly stable and has successfully recognized 277 samples. The algorithm also works for small-scale or weak gust fronts. 4） Radar data quality has certain impact on the algorithm.