多因变量双因子集矩阵回归预报方法

在多因变量多元回归中,采用单因子筛选引入公共预报因子进行矩阵回归预报时,由于自变量间互相关联,有可能存在这样的双因子,单个因子回归检验不太显著,但2个因子检验变得显著,势必造成某些预报信息的遗漏。基于此该文引入多因变量双因子集矩阵回归预报方法,通过对玉溪市9站点5月雨量预报检验,预报效果明显高于单因子筛选回归,尤其对特多特少异常年更具有指导意义。     

东营市相对湿度集成预报系统

运用多元回归法和多因子权重回归法分别建立预报模型，并在此基础上建立了相对湿度集成预报系统。     

东营市相对湿度集成预报系统

运用多元回归法和多因子权重回归法分别建立预报模型,并在此基础上建立了相对湿度集成预报系统.     

SVM方法在热带气旋降水预报中的应用

应用国家气象中心下发的热带气旋报文资料（中心经纬度、中心气压、近中心最大风力）,设计了诸如热带气旋中心与嘉兴市的最近距离等因子,并任选3个因子组成三维坐标（x,y,z）,计算嘉兴市热带气旋降水的概率分布函数。利用SVM回归方法对选中的概率分布函数进行训练,建立嘉兴市热带气旋降水预报方法,并应用于业务试报。同时和多元回归方法进行历史回报和业务试验比较。结果表明,概率分布函数与热带气旋降水具有较好的相关性,SVM回归方法在嘉兴市热带气旋降水预报中的回归预报模型和业务化效果明显优于多元回归。     

县站温度集成预报方法

介绍了以各台、站气温预报为因子，县站最高、最低气温为预报对象，用多元回归技术做温度集成预报的方法。     

多层递阶预报模型试做长期天气预报

根据预报量与预报因子的实际情况，提出用灰色系统中的关联分析筛选非线性因子。同时利用时变参数跟踪理论，将复杂的非线性系统预报问题转化成简单的带时变参数的线性多层递阶预报问题。经实际应用证明，用此法建立的多层递阶预报模型优于线性因子所建立的多层递阶预报模型。     

用多元线性回归分析作凯里站5月逐日晴雨预报

利用多元回归分析方法建立凯里站逐日晴雨预报方程。预报因子的资料来自“黔东南州短期预报业务自动化系统”，通过“0，1”编码方案进行处理。建立的回归方程预报效果呆投入业务使用。     

区间概率函数在晋城市降水概率预报中的应用研究

本文主要应用区间概率函数转换预报因子，经过相关比筛选。进行因子回归，构建分量级降水概率预报方程，同时应用T213L31数值预报产品统计释用方法，结合最优子集回归构造分时刻预报方程，最后进行概率预报。我们根据晋城气候区划，设计平川和山区两种降水概率预报方程。为晋城市开展降水概率预报提供了基础。     

基于EC025配料的短时强降水概率预报产品及应用

利用ECMWF预报资料,从动力、热力、水汽、能量和降水预报5个方面选取影响短时强降水发生的因子,构建多因变量数组,并利用主成分分析确定配料系数及其阈值,在此基础上进行配料,研发了四川省短时强降水概率预报产品投入应用。结果表明:产品对盆地20mm/h以上和高原地区10mm/h以上的短时强降水落区预报效果显著。      

分县客观定量降水预报微机自动化处理系统

本系统着重于T106数据预报产品的应用，采用最优子集回归方程，从气象卫星网上下发的T106数值预报产品中选取最优因子，建立了长治地区11个县市24h、48h降水预报多元回归方程。该系统采用VB5编程，建立了实时资料自动处理、分县客观定量降水预报自动制作、预报结果文本、图形输出等模块。该系统可直接挂接于MICAPS系统下，定时自动运行，无需人工干预。预报结果输出图文并茂，可供预报员参考。     

西北太平洋热带气旋强度统计释用预报方法研究

为了提高西北太平洋地区热带气旋（TC）强度预报准确率，在气候持续预报方法基础上，考虑气候持续性因子、天气因子、卫星资料因子，以TC强度变化为预报对象，运用逐步回归统计方法，建立西北太平洋地区24、48、72小时TC强度预报方程。通过不同的分海区试验（远海区域、华东近海、华南近海），证明回归结果较好。逐一分析选入因子发现：气候持续性因子在方程中相当重要；同时对远海区域和华东近海而言，海温影响也不容忽视，对华南近海而言，反映动力强迫作用的因素也较为重要。卫星资料的加入，对回归结果略有改进。用“刀切法”作独立样本检验，与气候持续法比较，预报误差明显减小。     

Computation of the wind speed at sites with complex orographic effects

Computation of the wind speed at a given station is performed via interpretation of the hydrodynamic modelling results using the multiple regression method. The a priori model parameters—pressure and thermal gradients, the Laplace operator at the standard levels of the lower and middle troposphere, and others—were obtained from reanalysis. The wind speed at observation times at Vladivostok station was used as a dependent variable. Preliminary classifying of the vector of predictors in accordance to the surface pressure gradient has been done. This procedure, first of all, allowed us to decrease variability of the friction force impact on the wind speed formation for a given class. Experiments, which were performed with the different sets of parameters and various classification schemes, allowed selecting models that provide computation results comparable to the known methods of wind speed computation at a given station.     

Using the quantile regression method to analyze changes in climate characteristics

Possibilities to use the non-parametric regression analysis method, named the quantile regression, for the estimation of changes in climate characteristics are considered. When analyzing the trends of climatic series, the quantile regression method enables to get the information on trends along the whole range of quantile values from 0 to 1 of dependent variable distributions, that is more informative than the use of traditional regression technique, based on the least-squares method (LSM) and enabling to obtain trend estimations for average values of the dependent variable only. Trend estimation errors for various methods are analyzed. The computation of quantile regression parameters for real climatic series is executed. Series of meteorological variables of the diurnal resolution, which characterize the surface climate (minimal, average, and maximal diurnal temperatures) and free atmosphere climate (temperature of isobaric surfaces up to 30 hPa inclusive) are considered. Seasonal peculiarities in trend manifestation at different parts of quantile range of these meteorological values are discussed. Concerning the problem of the analysis of climate trends, the quantile regression method seems to be perspective from the point of view of more detailed understanding of processes in the climate system, such as the surface and tropospheric warming, stratospheric cooling, long-period changes in characteristics of climate variability and extremity.     

统计学长期天气预报方法的若干研究(一) 逐步回归技术的应用

本文采用了逐步迴归技术,可以在一大堆预报因子中选择出迴归方程中的主要变量,组成一个最后的迴归方程——预报方程。 为了具体说明逐步迴归方法应用于长期预报的计算情况,将1932—1962年华北五站(北京、天津、保定、石家庄、营口)7,8月份平均降水总量的资料作为因变量y,以表征太阳辐射影响的因子、前期环流影响的因子和其它若干气象要素等作为自变量进行分析研究。 对我国若干重点地区的月、季降水的迴归分析指出,各地区与降水有关的预报因子是不完全相同的。就1963年夏季和6—10月份月的降水长期预报进行检查。各重点地区42次预报的结果说明,用这个方法所作的降水趋势预报尚好,比偶然性预报约高10％。 最后,作者就迴归分析的改进作了简略的讨论,并提出今后除了用于长期预报的研究和业务工作外,对中、短期和专业天气预报以及气候学方面等的研究工作也将会是一个有用工具。     

Statistical downscaling of historical monthly mean winds over a coastal region of complex terrain. I. Predicting wind speed

Surface wind speed is a key climatic variable of interest in many applications, including assessments of storm-related infrastructure damage and feasibility studies of wind power generation. In this work and a companion paper (van der Kamp et al. 2011), the relationship between local surface wind and large-scale climate variables was studied using multiple regression analysis. The analysis was performed using monthly mean station data from British Columbia, Canada and large-scale climate variables (predictors) from the NCEP-2 reanalysis over the period 1979–2006. Two regression-based methodologies were compared. The first relates the annual cycle of station wind speed to that of the large-scale predictors at the closest grid box to the station. It is shown that the relatively high correlation coefficients obtained with this method are attributable to the dominant influence of region-wide seasonality, and thus contain minimal information about local wind behaviour at the stations. The second method uses interannually varying data for individual months, aggregated into seasons, and is demonstrated to contain intrinsically local information about the surface winds. The dependence of local wind speed upon large-scale predictors over a much larger region surrounding the station was also explored, resulting in 2D maps of spatial correlations. The cross-validated explained variance using the interannual method was highest in autumn and winter, ranging from 30 to 70% at about a dozen stations in the region. Reasons for the limited predictive skill of the regressions and directions for future progress are reviewed.     

定性和定量长期预报模型的综合分析

根据相同的 50 0 h Pa和海温场预报因子 ,利用神经网络灵活可变的拓朴结构 ,分别构造了定性和定量的降水长期预报模型。并在同等条件下 ,建立了逐步回归预报方程。通过对比分析表明 ,这种定性和定量相结合的神经网络综合预报分析方法 ,是增强预报结果可靠性和稳定性的一种有效途径。该预报建模方法具有比较合理的分析依据 ,值得进一步探索、应用。     

基于主成分分析的人工智能台风路径预报模型

利用主成分分析可以从具有随机噪声干扰的气象场提取主要信号特征,排除随机干扰的能力,论文以1980～2010年共31年6～9月西行进入南海海域的台风样本为基础,综合考虑台风移动路径的气候持续因子和数值预报产品动力预报因子,采用主成分分析的特征提取与逐步回归计算相结合的预报因子信息数据挖掘技术,以进化计算的遗传算法,生成期望输出相同的多个神经网络个体,建立了一种新的非线性人工智能集合预报模型,进行了分月台风路径预报模型的预报建模研究。在预报建模样本、独立预报样本相同的情况下,分别采用人工智能集合预报方法和气候持续法进行了预报试验,试验对比结果表明,前者较后者在6、7、8和9月份台风路径预报中,平均绝对误差分别下降了7.4%、4.8%、12.4%、17.0%。另外,论文进一步在初选预报因子和样本个例相同的情况下,通过比较新模型与直接采用主成分分析方法选因子并分别运用逐步回归和遗传—神经网络集合预报模型进行计算的预报精度差异表明,前者具有更高的预报精度,其原因是该方法挖掘利用了全部备选预报因子的有用预报信息,而且遗传—神经网络集合预报模型的是由多个神经网络个体预报结果合成,集合模型的各个神经网络个体的网络结构,是通过遗传算法的优化计算确定的,因此,该集合预报模型的泛化能力显著提高,在实际天气预报中具有较好的实用性和推广价值。     

Quadratic discrimination through orthogonal transformation and its application to long-range forecasting of drought and excessive rainfall

The method of quadratic discrimination through orthogonal transformation is introduced to optimize quadratic discriminant function, which has been proved more effective than the method of stepwise multiple discrimination. It is noteworthy that, in the case of a large number of predictors, it is better to make a preliminary choice by using stepwise multiple discrimination or stepwise regression so that the calculation can be made more stable and the effectiveness of discrimination can be improved.     

Compilation method for 1 km grid data of monthly mean air temperature for quantitative assessments of climate change impacts

A new method is proposed to compile 1 km grid data of monthly mean air temperature by dynamically downscaling general circulation model (GCM) data with a regional climate model (RCM). The downscaling method used is a technique referred to as the pseudoglobal warming method to reduce GCM bias. For the grid data, RCM data were corrected with data from an existing meteorological network. The correction model for the RCM bias was developed by stepwise multiple regression analysis using the difference in the monthly mean air temperatures between the observation and RCM output as a dependent variable and the geographical factors as independent variables. Our method corrected the RCM bias from 1.69°C to 0.58°C for the month of August in the 1990s (1990–1999).     

长期预报基本观测资料自动化处理

将微型气象无人驾驶飞机与无线数字影像采集、传输设备相结合,构成了一套微型无人机空中火情监测系统,并开发了相应的地面监测软件.经过地面测试和空中飞行试验证实,该系统运行良好,对于地面火点有较好的探测能力.目前试验系统已具备单架次完成半径30 km以内,面积80～100 km2林区巡护任务的能力,而且这一系统成本低、使用灵活方便,具有良好的应用前景.同时就系统开发中遇到的相关问题和今后的改进方向进行了讨论.