用于ENSO预测的一种广义典型混合回归模式及其预报试验

提出一种基于主分量典型相关分析（ＰＣ－ＣＣＡ）的广义典型混合回归模式，用于建立ＮＩＮＯ海区ＳＳＴ预报方案．该模式引入ＥＥＯＦ、ＰＲＥＳＳ准则和集成预报等技术思想，在优选物理因子，确定最佳模式参数的基础上，对ＮＩＮＯ海区海温指数所作的超前１—４季度预报试验取得优良效果．试验表明，该模式方案性能稳定，其总体预报技术水平已达到美国ＮＯＡＡ／ＮＷＳ／ＮＣＥＰ／气候诊断公报（ＣＰＣ）所用同类模式水平．而本模式方案预报同类产品所需因子数远少于ＣＰＣ方法。这就有可能为建立我国的ＥＮＳＯ业务监测系统提供有益的基础。     

基于神经网络奇异谱分析的ENSO指数预测

采用奇异谱分析方法研究了ENSO指数及相关序列, 结果表明奇异谱分析能很好的对原始序列进行信噪分离, 增大了ENSO指数的可预报性.在此基础上, 提出了人工神经网络和奇异谱分析相结合的ENSO指数预测方法, 进行了不同因子组合的预报试验, 预报效果明显优于持续性预报, 超前4季的Nio 3区、 Nio 4区预报相关系数仍高于0.5.     

全球热带简单海气耦合模式中的ENSO预报试验

利用一个全球热带简单海气耦合模式(GTSM模式),并选取热带三大洋较强的冷暖事件作为预报对象进行了若干预报试验,分析结果发现:在GTSM模式中由于热带三大洋海气耦合通过大气模式而相互作用和影响,使得该模式对于东大西洋和中东印度洋较强冷暖事件的预报能力,较单独大西洋或单独印度洋耦合模式均有明显提高,预报和观测的ATL3、IND3指数的相关系数达到0.5以上的月份,分别达到9个月和6个月左右;而在东太平洋则和ZC(LDEO1)模式差不多,预报和观测的Nio3指数的相关系数达到0.6以上的月份可以达到15个月左右.     

关于ENSO及其统计预测方法评估

结合近年来“九五”国家重中之重项目有关ENSO业务预报系统研究工作的体会和见解 ,简要综述了近年来ENSO的预报问题 ,尤其是ENSO统计预测方法的研究成果 ,并对各种预报方法的效果进行评价。     

混合海气耦合模式中的ENSO循环及其形成机制

在无异常外强迫的情况下, 将混合海气耦合模式进行了45年的模拟积分.结果表明:模式能较好地再现类似ENSO循环的热带太平洋海洋、大气的年际振荡, 模式ENSO循环的主周期为4~5年; 探讨了ENSO循环的负反馈机制, 指出:暖态的消亡与El Niño发展过程中太平洋东部不断增强的东风异常所产生的冷水上翻的加强以及纬向向西的冷平流有关; 冷态的消亡主要由赤道波的时滞效应所致.     

简化ENSO预测模式的改进试验

利用NCEP的海洋模式同化资料,对Cane-Zebiak简化海气耦合模式的海洋动力要素进行对比分析.结果表明,该模式的大气部分能够较好地模拟出海温强迫下的大气风场异常,而海洋部分的模拟结果与实况相比存在较多虚假的偏暖偏冷事件以及对El Nino事件的模拟强度偏小等问题.这主要与海洋模式中混合层厚度选取较浅,导致垂直上翻温度平流的贡献削弱有关.试验结果证明,适当加强该项的强度可提高模式的预报能力.     

一个可供ENSO预测的海气耦合环流模式及1997/1998 ENSO的预测

利用中国科学院大气物理研究所设计发展的具有较高分辨率的热带太平洋和全球大气耦合环流模式,设计了一个初始化方案,建立了ＥＮＳＯ预测系统,进行了系统性的预测试验。预测结果检验评估表明,该预测系统表现出较强的预报能力,赤道中东太平洋地区（Ｎｉｎｏ３和Ｎｉｎｏ３４）海表温度距平预报相关技巧高于０５２的预报可持续１８个月,该预测系统可应用到试验性的海温预测实践中。利用该系统对１９９７／１９９８年ＥＮＳＯ进行了实际预测,表明预测是成功的,预测的海温距平已提供给今年我国夏季降水预测使用,取得了良好的预测效果。     

全球气温与ENSO多年际耦合振荡关系的初步研究

借助于最优分割法和多通道奇异谱分析（MSSA）等方法对全球气候变化背景场阶段性变化及其与ENSO振荡信号的相互关系作了详细的诊断分析。结果表明：（1）全球近百年气温长期变化阶段性明显,各阶段之间的气候变率特征差异较大;（2）全球气温变化背景场的阶段性对ENSO的年际及年代际准周期振荡有较为明显的影响;（3）全球气温与Nino区海温的年际变率存在着准4年和准2年的显著耦合振荡。     

模式气候吸引子信息约束下的动力协调初始化方法及其在ENSO预测中的应用

为了改善模式初始场质量,减少初值与模式不协调对ENSO预测的影响,针对国家气候中心NCCo海-气耦合模式原初始化方案动力小协调的问题,从利用模式长期耦合模拟资料中的模式气候吸引子信息的角度出发,发展了一种获取观测资料中与模式相协调分量的信息重构方法,提出了一种模式气候吸引子信息约束下的动力协调初始化方案.对该方案回报检验的结果表明:通过反演NCCo海-气耦合模式模拟资料中的模式气候吸引子信息,有助于获取观测资料中与模式相协调的信息分量特征,实现了初始化过程中动力模式与所同化观测资料间的协调.这种基于信息重构方法的动力协调初始化方案,既可以延续原初始化方案利用观测信息较多的优势,又克服了原方案中观测资料和动力模式不协调的缺陷.这种新的初始化方案,消除了观测资料和模式不协调在初始场中产生的小尺度高频噪声,突出了与NCCo模式动力特征相适应的ENSO尺度信息.进而抑制了初始场中高频噪声所引起的快变预报误差的增长,提高了模式的预测技巧.     

A GENERALIZED CANONICAL MIXED REGRESSION MODEL FOR ENSO PREDICTION WITH ITS EXPELMENT

A scheme is proposed for predicting NINO-region SST in terms of a generalized canonical mixedregression model based on principal component canonical correlation analysis(PC-CCA). and into the scheme are introduced such techniques as EEOF, PRESS criterion and consensus prediction. By optimizing physicalfactors and selecting optimal model parameters, experiments were made successful in predicting the LINO SST index for 1 to 4 seasons to follow. The scheme is shown to be stable in operation and its total technical level compares well with that of the model published in NOAA/NWS/NCEP CPC Climate Diagnostics Bulletin. butthe number of factors needed in our scheme is much fewer than that for the CPC's model in dealing with the sameproblems. This makes it possible to establish an operational ENSO monitoring system in China.     

Nino海区SSTA短期气候预测模型试验

采用奇异谱分析（ＳＳＡ）与自回归（ＡＲ）预测模型相结合的方案,对Ｎｉｎｏ海区平均ＳＳＴ逐月距平序列作自适应滤波意义下的超前预报。结果表明,对１９９７－１９９８年这次强ＥＮＳＯ事件的超前预报十分有效;利用相应的历史样本作三次强ＥＮＳＯ事件的回溯预报试验,发现均有较高可信度。可见,该方案预报技巧稳定,独立样本试验和实际预报试验都有很高的准确率。将ＳＳＡ－ＡＲ方案进一步完善,可望作为ＥＮＳＯ业务预报的有     

STUDY ON MIXED MODEL OF NEURAL NETWORK FOR FARMLAND FLOOD/DROUGHT PREDICTION

The paper concerns a flood/drought prediction model involving the continuation of time seriesof a predictand and the physical factors influencing the change of predictand.Attempt is made toconstruct the model by the neural network scheme for the nonlinear mapping relation based onmulti-input and single output.The model is found of steadily higher predictive accuracy by testingthe output from one and multiple stepwise predictions against observations and comparing theresults to those from a traditional statistical model.     

CLIMATE PREDICTION EXPERIMENT FOR TROPICAL CYCLONE GENESIS FREQUENCY USING THE LARGE-SCALE CIRCULATION FORECAST BY A COUPLED GLOBAL CIRCULATION MODEL

Based on an analysis of the relationship between the tropical cyclone genesis frequency and large-scale circulation anomaly in NCEP reanalysis, large-scale atmosphere circulation information forecast by the JAMSTEC SINTEX-F coupled model is used to build a statistical model to predict the cyclogenesis frequency over the South China Sea and the western North Pacific. The SINTEX-F coupled model has relatively good prediction skill for some circulation features associated with the cyclogenesis frequency including sea level pressure, wind vertical shear, Intertropical Convergence Zone and cross-equatorial air flows. Predictors derived from these large-scale circulations have good relationships with the cyclogenesis frequency over the South China Sea and the western North Pacific. A multivariate linear regression （MLR） model is further designed using these predictors. This model shows good prediction skill with the anomaly correlation coefficient reaching, based on the cross validation, 0.71 between the observed and predicted cyclogenesis frequency. However, it also shows relatively large prediction errors in extreme tropical cyclone years （1994 and 1998, for example）.     

A STATISTICAL PREDICTION MODEL FOR SPRING RAINFALL OVER NORTHERN CHINA BASED ON THE INTERANNUAL INCREMENT APPROACH

The spring (March-April-May) rainfall over northern China (SPRNC) is predicted by using the interannual increment approach. DY denotes the difference between the current year and previous years. The seasonal forecast model for the DY of SPRNC is constructed based on the data that are taken from the 1965-2002 period (38 years), in which six predictors are available no later than the current month of February. This is favorable so that the seasonal forecasts can be made one month ahead. Then, SPRNC and the percentage anomaly of SPRNC are obtained by the predicted DY of SPRNC. The model performs well in the prediction of the inter-annual variation of the DY of SPRNC during 1965-2002, with a correlation coefficient between the predicted and observed DY of SPRNC of 0.87. This accounts for 76% of the total variance, with a low value for the average root mean square error (RMSE) of 20%. Both the results of the hindcast for the period of 2003-2010 (eight years) and the cross-validation test for the period of 1965-2009 (45 years) illustrate the good prediction capability of the model, with a small mean relative error of 10%, an RMSE of 17% and a high rate of coherence of 87.5% for the hindcasts of the percentage anomaly of SPRNC.     

NINO区SST与SOI的耦合振荡信号及其预测试验

应用奇异交叉谱（ＳＣＳＡ）分析方法,提取Ｎｉｎｏ 海区各区的平均海温（ＳＳＴ）和南方涛动指数（ＳＯＩ）之间的耦合振荡信号,由此描述其年际和年代际的时变特征。基于ＳＣＳＡ,重建耦合振荡分量序列（ＲＣＣＳ）,并与回归分析相结合,对Ｎｉｎｏ 各海区平均的ＳＳＴ月际序列作短期气候预测试验。结果表明,各海区ＳＳＴ与ＳＯＩ的显著耦合振荡周期各有特色,其年际或１０ 年际变化不尽相同,从而构成了ＥＮＳＯ信号在时空演变型态上的复杂性。ＳＣＳＡ基础上的回归预报模型的预报技巧绝大部分优于ＳＳＡ－ＡＲ预报模型,实际预报试验证明效果优良     

利用PRESS准则和岭回归方法建立大气遥感最优反演方程

在文献［１］的基础上，首次引用近代回归分析中ＰＲＥＳＳ准则最优于集法，建立大气温湿遥感最优反演方程，并用岭回归分析评价不同准则或方法所得到的最优子集，同时用实际资料进行了验证。     

A PREDICTION SCHEME FOR THE PRECIPITATION OF SPR BASED ON THE DATA MINING ALGORITHM AND CIRCULATION ANALYSIS

Based on the 74 circulation indexes provided by National Climate Center of China (hereinafter referred to as NCC) and the 24 indexes compiled by NOAA, the study used the C4.5 algorithm in data mining to establish a decision tree prediction model to predict whether the Spring Persistent Rains (hereinafter referred to as SPR) of 55 years (from 1961 to 2015) is more than the normal, and obtained 5 rules to determine whether the SPR is more than the normal. The accuracy rate of the test set, namely “whether the SPR is more than the normal”, is 98.18%. After evaluating the model by conducting ten 10-fold cross validations to take the average value, the test accuracy rate gained is 84%. There are differences between the three types of years with a SPR more than the normal when it comes to intensity and distribution. In spring, they have respective anomalous 850hPa monthly mean wind fields and water-vapor flux distribution, and 700hPa forms the zone where the vertical speed is anomalously negative. As indicated by the results, the SPR prediction model based on the C4.5 algorithm has a high prediction accuracy rate, the model is reasonably and effectively constructed, and the decision rules take comprehensive factors into consideration. The anomalous rainfall and circulation distribution characteristics obtained based on the decision classification results provide new ideas and methods for the climatic prediction of SPR.