主成分估计在多项式曲面拟合GPS高程中的应用

阐述了多项式曲面拟合GPS高程中参数数目对结果的影响,比较了传统最小二乘平差、主成分估计和岭估计在不同参数时的计算结果,主成分回归在不同参数时都可以取得比较满意的结果。     

地图数字化中基于点位坐标的统一平差模型

探讨了地图数字化过程中采用平差方法来建立坐标计算的统一模型。将测得的角度、边长等数据化为方向、距离观测值列立误差方程，通过调用统一的平差迭代计算程序来计算未知点坐标及精度。详细给出了基于平差的11种常用坐标计算方法，并通过实例进行了说明。     

闽中北前汛期多普勒雷达定量估测降水分析

使用2002、2003年福建中北部27个气象站5—6月自记雨量资料及相应的雷达体扫资料，通过用最优化法统计得出不同区域的Z-I关系，并将统计结果用于2005年5—6月的降水估测。同时利用实时雨量资料采用卡尔曼最优（卡尔曼滤波＋最优插值）、变分等估测方法对2005年几次混合型降水过程进行实时雨量校正，用福建北部武夷山九曲溪流域雨量计检验校正后的雨量值，并对上述几种方法的测量精度进行比较、分析。结果表明：卡尔曼最优、变分法及100km距离范围内的最优化法的过程平均相对误差小于25％，100km距离范围以外最优化法的过程平均相对误差明显增大，Z=300I^1,4估测的误差最大。     

登陆台风短时定量降水预报方法初探

利用登陆台风的GMS-5 IR1 TBB资料及地面自动站逐时雨量资料,初步建立了适用于登陆台风的定量降水估计(QPE)方法,基于QPE结果,采用外推法,初步实现了对登陆台风未来0~3 h的短时定量降水预报(QPF)。应用于0104号登陆台风“尤特”和0414号登陆台风“云娜”两个例表明:对单站逐小时雨量定量预报而言,如果50%相对误差可以接受,01、、23、h的QPF准确率分别达20%~70%、30%~80%、30%~70%、20%~60%;对于单站过程雨量定量预报而言,如果50%相对误差可以接受,0、1、2、3 h的QPF准确率基本都达60%~80%;对未来0~3 h的面雨量也具有一定定量预报能力。0~3 h QPF结果对实际短时登陆台风降水业务预报工作具有一定的参考价值。     

全球资料同化中误差协方差三维结构的准确估计与应用Ⅱ:背景误差协方差调整与数值试验分析

利用NMC方法针对背景误差协方差的方差、三维相关与特征长度来揭示T213L31模式的误差主要特点,并与传统更新矢量方法的计算结果进行了对比与调整。结果表明NMC方法结果与更新矢量方法结果在大体特征上基本吻合,但细节上的差异不可忽视,特别是对背景误差方差与特征长度的估计存在显著的差异,其主要原因是NMC方法倾向于高估天气尺度波的背景误差,而低估次天气尺度到中尺度波的背景误差。通过对背景误差方差、特征长度的调整,显著改善了背景误差功率谱的分布特点,使得NMC方法结果与更新矢量方法结果更为吻合。通过三维变分同化与最优插值中观测与背景误差相对重要性的比较,发现两者结果基本一致,但三维变分同化在850 hPa以下的温度场和300 hPa以上的风场统计结果都表现出背景误差相对于观测误差偏小的特点。背景误差相对于观测误差偏小有助于保证分析场中质量场与风场平衡,消除了大气底层和高层质量场与风场不匹配现象。在数值试验中,针对不同的背景误差均方差与特征长度的特点,分析了分析增量和预报效果的差异,结果表明,准确的背景误差估计与优化工作改善了预报效果,使得北半球三维变分同化的120 h预报效果整体好于现有最优插值。     

An Aggregate Constraint Method for Inequality-constrained Least Squares Problems

The inequality-constrained least squares (ICLS) problem can be solved by the simplex algorithm of quadratic programming. The ICLS problem may also be reformulated as a Bayesian problem and solved by using the Bayesian principle. This paper proposes using the aggregate constraint method of non-linear programming to solve the ICLS problem by converting many inequality constraints into one equality constraint, which is a basic augmented Lagrangean algorithm for deriving the solution to equality-constrained non-linear programming problems. Since the new approach finds the active constraints, we can derive the approximate algorithm-dependent statistical properties of the solution. As a result, some conclusions about the superiority of the estimator can be approximately made. Two simulated examples are given to show how to compute the approximate statistical properties and to show that the reasonable inequality constraints can improve the results of geodetic network with an ill-conditioned normal matrix.     

A preliminary investigation of strong-motion data from the French Antilles

Strong-motion networks have been operating in the Caribbean region since the 1970s, however, until the mid-1990s only a few analogue stations were operational and the quantity of data recorded was very low. Since the mid-1990s, digital accelerometric networks have been established on islands within the region. At present there are thought to be about 160 stations operating in this region with a handful on Cuba, 65 on the French Antilles (mainly Guadeloupe and Martinique), eight on Jamaica, 78 on Puerto Rico (plus others on adjacent islands) and four on Trinidad.After briefly summarising the available data from the Caribbean islands, this article is mainly concerned with analysing the data that has been recorded by the networks operating on the French Antilles in terms of their distribution with respect to magnitude, source-to-site distance, focal depth and event type; site effects at certain stations; and also with respect to their predictability by ground motion estimation equations developed using data from different regions of the world. More than 300 good quality triaxial acceleration time-histories have been recorded on Guadeloupe and Martinique at a large number of stations from earthquakes with magnitudes larger than 4.8, however, most of the records are from considerable source-to-site distances. From the data available it is found that many of the commonly-used ground motion estimation equations for shallow crustal earthquakes poorly estimate the observed ground motions on the two islands; ground motions on Guadeloupe and Martinique have smaller amplitudes and are more variable than expected. This difference could be due to regional dependence of ground motions because of, for example, differing tectonics or crustal structures or because the ground motions so far recorded are, in general, from smaller earthquakes and greater distances than the range of applicability of the investigated equations.     

基于虚拟观测的病态问题解法

在大地测量数据处理领域中,处理病态问题的主要方法有:岭估计方法、奇异值分解法(SVD)、Tik-honov正则化方法等,但是这些方法大多数是强调数学上的意义,没有充分联系大地测量的实际情况,因此不利于在测绘领域病态问题本质的理解和研究。为使病态问题的求解具有实际的物理意义,提出了基于虚拟观测的岭估计方法。该方法将先验约束条件作为一类互相独立的虚拟观测值,从而把病态问题转化为测量平差问题,然后运用Helmert方差估计法来确定岭参数。该方法还可以得到的参数之间的权矩阵,用它来代替虚拟观测值的权矩阵,重新对参数进行计算,则实现了该方法向广义岭估计的推广。实际算例分析的结果表明该方法不仅计算简单而且能保证结果精确。     

A radar-based verification of precipitation forecast for local convective storms

Local flash flood storms with a rapid hydrological response are a real challenge for quantitative precipitation forecasting (QPF). It is relevant to assess space domains, to which the QPF approaches are applicable. In this paper an attempt is made to evaluate the forecasting capability of a high-resolution numerical weather prediction (NWP) model by means of area-related QPF verification. The results presented concern two local convective events, which occurred in the Czech Republic (CR) on 13 and 15 July 2002 and caused local flash floods. We used the LM COSMO model (Lokall Model of the COSMO consortium) adapted to the horizontal resolution of 2.8 km over a model domain covering the CR. The 18 h forecast of convective precipitation was verified by using radar rainfall totals adjusted to the measured rain gauge data. The grid point-related root mean square error (RMSE) value was calculated over a square around the grid point under the assumption that rainfall values were randomly distributed within the square. The forecast accuracy was characterized by the mean RMSE over the whole verification domain. We attempt to show a dependence of both the RMSE field and the mean RMSE on the square size. The importance of a suitable merger between the radar and rain gauge datasets is demonstrated by a comparison between the verification results obtained with and without the gauge adjustment. The application of verification procedure demonstrates uncertainties in the precipitation forecasts. The model was integrated with initial conditions shifted by 0.5° distances. The four verifications, corresponding to the shifts in the four directions, show differences in the resulting QPF, which depend on the size of verification area and on the direction of the shift.     

预测矿井涌水量的计算级别与精度评述

根据水利部2005年发布的《建设项目水资源论证导则（试行）》技术文件中的有关规定与要求,参照《供水水文地质勘察规范》以及《固体矿产资源/储量分类,》认为地下水可开采量与预测矿井涌水量的计算可按勘查研究程度,将其可分为A、B、C、D、E等5个精度等级。参照各级别的允许误差,论证了目前矿井涌水量预测计算公式及方法的可靠程度。