基于海气耦合环流模式的ENSO预测

Predictions of ENSO are described by using a coupled atmosphere-ocean general circulation model. The initial conditions are created by forcing the coupled system using SST anomalies in the tropical Pacific at the background of the coupled model climatology. A series of 24-month hindcasts for the period from November 1981 to December 1997 are carried out to validate the performance of the coupled system. Correlations of SST anomalies in the Nino3 region exceed 0.54 up to 15 months in advance and the rms errors are less than 0.9℃. The system is more skillful in predicting SST anomalies in the 1980s and less in the 1990s. The model skills are also seasonal-dependent, which are lower for the predictions starting from late autumn to winter and higher for those from spring to autumn in a year-time forecast length. The prediction, beginning from March, persists 8 months long with the correlation skill exceeding 0.6, which is important in predictions of summer rainfall in China. The predictions are succesful in many aspects for the 1997-2000 ENSO events.     

Time-Expanded Sampling Approach for Ensemble Kalman Filter:Experiment Assimilation of Simulated Soundings

In the Ensemble Kalman Filter (EnKF) data assimilation-prediction system, most of the computation time is spent on the prediction runs of ensemble members. A limited or small ensemble size does reduce the computational cost, but an excessively small ensemble size usually leads to filter divergence, especially when there are model errors. In order to improve the efficiency of the EnKF data assimilation-prediction system and prevent it against filter divergence, a time-expanded sampling approach for EnKF based on the WRF (Weather Research and Forecasting) model is used to assimilate simulated sounding data. The approach samples a series of perturbed state vectors from Nb member prediction runs not only at the analysis time (as the conventional approach does) but also at equally separated time levels (time interval is △t) before and after the analysis time with M times. All the above sampled state vectors are used to construct the ensemble and compute the background covariance for the analysis, so the ensemble size is increased from Nb to Nb+2M￡Nb=(1+2M)×Nb) without increasing the number of prediction runs (it is still Nb). This reduces the computational cost. A series of experiments are conducted to investigate the impact of △t (the time interval of time-expanded sampling) and M (the maximum sampling times) on the analysis. The results show that if △t and M are properly selected, the time-expanded sampling approach achieves the similar effect to that from the conventional approach with an ensemble size of (1+2M)×Nb, but the number of prediction runs is greatly reduced.     

The 3-hour-interval prediction of ground-level temperature in South Korea using dynamic linear models

The 3-hour-interval prediction of ground-level temperature from 00 h out to 45 h in South Korea(38 stations) is performed using the DLM (dynamic linear model) in order to eliminate the systematic error of numerical model forecasts. Numerical model forecasts and observations are used as input values of the DLM. According to the comparison of the DLM forecasts to the KFM (Kalman filter model) forecasts with RMSE and bias, the DLM is useful to improve the accuracy of prediction.     

Probabilistic Flood Prediction in the Upper Huaihe Catchment Using TIGGE Data

Based on the precipitation and temperature data obtained from THORPEX (The Observing System Research and Predictability Experiment) Interactive Grand Global Ensemble (TIGGE)-China Meteorological Administration (CMA) archiving center and the raingauge data,the three-layer variable infiltration capacity (VIC-3L) land surface model was employed to carry out probabilistic hydrological forecast experiments over the upper Huaihe River catchment from 20 July to 3 August 2008.The results show that the performance of the ensemble probabilistic prediction from each ensemble prediction system (EPS) is better than that of the deterministic prediction.Especially,the 72-h prediction has been improved obviously.The ensemble spread goes widely with increasing lead time and more observed discharge is bracketed in the 5th-99th quantile.The accuracy of river discharge prediction driven by the European Centre (EC)-EPS is higher than that driven by the CMA-EPS and the US National Centers for Environmental Prediction (NCEP)-EPS,and the grand-ensemble prediction is the best for hydrological prediction using the VIC model.With regard to Wangjiaba station,all predictions made with a single EPS are close to the observation between the 25th and 75th quantile.The onset of the flood ascending and the river discharge thresholds are predicted well,and so is the second rising limb.Nevertheless,the flood recession is not well predicted.     

一个全球耦合模式的ENSO后报试验

A group of seasonal hindcast experiments are conducted using a coupled model known as the Flexible Global Ocean-Atmosphere-Land System Modelgamil1.11 (FGOALS-g1.11) developed at the State Key Laboratory of Numerical Modeling for Atmospheric Sciences and Geophysical Fluid Dynamics (LASG).Two steps are included in our ElNi o-Southern Oscillation (ENSO) hindcast experiments.The first step is to integrate the coupled GCM with the Sea Surface Temperature (SST) strongly nudged towards the observation from 1971 to 2006.The second step is to remove the SST nudging term.The authors carried out a one-year hindcast by adopting the initial values from SST nudging experiments from the first step on January 1st,April 1st,July 1st,and October 1st from 1982 to 2005.In the SST nudging experiment,the model can reproduce the observed equatorial thermocline anomalies and zonal wind stress anomalies in the Pacific,which demonstrates that the SST nudging approach can provide realistic atmospheric and oceanic initial conditions for seasonal prediction experiments.The model also demonstrates a high Anomaly Correlation Coefficient (ACC) score for SST in most of the tropical Pacific,Atlantic Ocean,and some Indian Ocean regions with a 3-month lead.Compared with the persistence ACC score,this model shows much higher ACC scores for the Ni o-3.4 index for a 9-month lead.     

Seasonal Climate Prediction Models for the Number of Landfalling Tropical Cyclones in China

Two prediction models are developed to predict the number of landfalling tropical cyclones(LTCs) in China during June–August(JJA). One is a statistical model using preceding predictors from the observation, and the other is a hybrid model using both the aforementioned preceding predictors and concurrent summer large-scale environmental conditions from the NCEP Climate Forecast System version 2(CFSv2).(1) For the statistical model, the year-to-year increment method is adopted to analyze the predictors and their physical processes, and the JJA number of LTCs in China is then predicted by using the previous boreal summer sea surface temperature(SST) in Southwest Indonesia,preceding October South Australia sea level pressure, and winter SST in the Sea of Japan. The temporal correlation coefficient between the observed and predicted number of LTCs during 1983–2017 is 0.63.(2) For the hybrid prediction model, the prediction skill of CFSv2 initiated each month from February to May in capturing the relationships between summer environmental conditions(denoted by seven potential factors: three steering factors and four genesis factors) and the JJA number of LTCs is firstly evaluated. For the 2-and 1-month leads, CFSv2 has successfully reproduced these relationships. For the 4-, 3-, and 2-month leads, the predictor of geopotential height at 500 h Pa over the western North Pacific(WNP) shows the worst forecasting skill among these factors. In general, the summer relative vorticity at 850 h Pa over the WNP is a modest predictor, with stable and good forecasting skills at all lead times.     

FORECASTING THE MJO INDEX BASED ON SSA-AR

Experiments of forecasting daily bi-variate index of the tropical atmospheric Madden-Julian Oscillation (MJO) are performed in the context of adaptive filtering prediction models by combining the singular spectrum analysis (SSA) with the autoregressive (AR) methods.the MJO index,a pair of empirical orthogonal function (EOF) time series,called RMM1 and RMM2,predicts by the combined statistical SSA and AR models:firstly,according to the index of historic observation decomposed by SSA and then reconstructed by selecting the first several components based on prominent variance contributions;after that,established an AR prediction model from the composite (scheme A) or built the forecast models for each of these selected reconstructed components,separately (Scheme B).Several experimental MJO index forecasts are performed based on the models.The results show that both models have useful skills of the MJO index forecast beyond two weeks.In some cases,the correlation coefficient between the observed and predicted index series stays above 0.5 in 20 leading days.The SSA-AR model,based on the reconstructed composite series,has better performance on MJO forecast than the AR model,especially for the leading time longer than 5 days.Therefore,if we build a real-time forecast system by the SSA-AR model,it might provide an applicable tool for the operational prediction of the MJO index.     

Wind-driven, double-gyre, ocean circulation in a reduced-gravity, 2.5-layer, lattice boltzmann model

A coupled lattice Boltzmann （LB） model with second-order accuracy is applied to the reduced-gravity, shallow water, 2.5-layer model for wind-driven double-gyre ocean circulation. By introducing the secondorder integral approximation for the collision operator, the model becomes fully explicit. The Coriolis force and other external forces are included in the model with second-order accuracy, which is consistent with the discretization accuracy of the LB equation. The feature of the multiple equilibria solutions is found in the numerical experiments under different Reynolds numbers based on this LB scheme. With the Reynolds number increasing from 3000 to 4000, the solution of this model is destabilized from the anti-symmetric double-gyre solution to the subtropic gyre solution and then to the subpolar gyre solution. The transitions between these equilibria states are also found in some parameter ranges. The time-dependent variability of the circulation based on this LB simulation is also discussed for varying viscosity regimes. The flow of this model exhibits oscillations with different timescales varying from subannual to interannual. The corresponding statistical oscillation modes are obtained by spectral analysis. By analyzing the spatiotemporal structures of these modes, it is found that the subannual oscillation with a 9-month period originates from the barotropic Rossby basin mode, and the interarmual oscillations with periods ranging from 1.5 years to 4.6 years originate from the recirculation gyre modes, which include the barotropic and the baroclinic recirculation gyre modes.     

A Short-Term Climate Prediction Model Based on a Modular Fuzzy Neural Network

In terms of the modular fuzzy neural network (MFNN) combining fuzzy c-mean (FCM) cluster and single-layer neural network, a short-term climate prediction model is developed. It is found from modeling results that the MFNN model for short-term climate prediction has advantages of simple structure, no hidden layer and stable network parameters because of the assembling of sound functions of the selfadaptive learning, association and fuzzy information processing of fuzzy mathematics and neural network methods. The case computational results of Guangxi flood season (JJA) rainfall show that the mean absolute error (MAE) and mean relative error (MRE) of the prediction during 1998 2002 are 68.8 mm and 9.78%, and in comparison with the regression method, under the conditions of the same predictors and period they are 97.8 mm and 12.28% respectively. Furthermore, it is also found from the stability analysis of the modular model that the change of the prediction results of independent samples with training times in the stably convergent interval of the model is less than 1.3 mm. The obvious oscillation phenomenon of prediction results with training times, such as in the common back-propagation neural network (BPNN) model, does not occur, indicating a better practical application potential of the MFNN model.     

Improving multi-model ensemble probabilistic prediction of Yangtze River valley summer rainfall

Seasonal prediction of summer rainfall over the Yangtze River valley(YRV) is valuable for agricultural and industrial production and freshwater resource management in China, but remains a major challenge. Earlier multi-model ensemble(MME) prediction schemes for summer rainfall over China focus on single-value prediction, which cannot provide the necessary uncertainty information, while commonly-used ensemble schemes for probability density function(PDF) prediction are not adapted to YRV summer rainfall prediction. In the present study, an MME PDF prediction scheme is proposed based on the ENSEMBLES hindcasts. It is similar to the earlier Bayesian ensemble prediction scheme, but with optimization of ensemble members and a revision of the variance modeling of the likelihood function. The optimized ensemble members are regressed YRV summer rainfall with factors selected from model outputs of synchronous 500-h Pa geopotential height as predictors. The revised variance modeling of the likelihood function is a simple linear regression with ensemble spread as the predictor. The cross-validation skill of 1960–2002 YRV summer rainfall prediction shows that the new scheme produces a skillful PDF prediction, and is much better-calibrated, sharper, and more accurate than the earlier Bayesian ensemble and raw ensemble.     

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     

A Multi-Statistical analysis of the southern oscillation (SO) and its relation to the mean monthly atmospheric circulation at 500 hPa in the northern hemisphere

In this paper the correlation analysis, factor analysis, fuzzy classification, and principal component analysis (PCA) are performed for the southern oscillation index (SOI) from the Climate Analysis Center (CAC) at the NOAA. It is shown that the 12-month SOI can be classified into two groups: one from January through April and the other from May through December. They differ in persistency and correlation. It is also found that the year of strong or weak SO can be defined by the first principal component of the SOI. The 11 years of weak SO thus defined contain 9 El Nino events.In addition, the relations between the SOI and 500 hPa geopotential height, mean monthly zonal height, mean monthly interzonal height differences, centers of atmospheric activities, characteristics of the atmospheric circulation (the intensity index of the north polar vortex, the area index of the subtropical West Pacific high, mean monthly zonal and meridional circulation indexes in Asia and Eurasia) in the period of 24 months from January through December of the next year have been examined on the basis of the monthly data from 1951 through 1984. The correlation coefficients and Mahalanobis distances are thus presented. Analysis indicates that in the early part of the low SOI year, i.e., in April, the 500 hPa geopotential height north of 75oN is significantly low and then becomes higher in May. It is found that in April the trough of the first harmonic wave is in the Eastern Hemisphere and the contribution of its variance is smaller than in May. Analysis shows that the opposite is true in the high SOI year. Such variation in the height field during the April-May period is an early signal of the SO at higher latitudes.In the end, a statistical prediction model for the SOI is presented, by means of which a low SOI year as well as an El Nino event has been successfully predicted for 1986.     

The Prediction of Non-stationary Climate Series Based on Empirical Mode Decomposition

This paper proposes a new approach which we refer to as ``segregated prediction" to predict climate time series which are nonstationary. This approach is based on the empirical mode decomposition method (EMD), which can decompose a time signal into a finite and usually small number of basic oscillatory components. To test the capabilities of this approach, some prediction experiments are carried out for several climate time series. The experimental results show that this approach can decompose the nonstationarity of the climate time series and segregate nonlinear interactions between the different mode components, which thereby is able to improve prediction accuracy of these original climate time series.     

Skill and reproducibility of seasonal rainfall patterns in the tropics in ECHAM-4 GCM simulations with prescribed SST

 The ECHAM4 atmospheric general circulation model (GCM) has been integrated at T30 resolution through the period 1960–1994 forced with the observed sea-surface temperatures (SSTs) as compiled at the Hadley Centre (GISST2.2). Three experiments were made starting from different initial conditions. The large-scale tropical precipitation patterns simulated by the model have been studied, focusing on the skill (i.e. the capability to simulate the observed anomaly over land areas) and reproducibility (i.e. the GCM’s interannual rainfall variance that is independent from the initial conditions). Analysis of variance is used to estimate the reproducibility amongst ensemble members at each grid-box, but most emphasis is placed on large-scale patterns, as revealed by various singular value decomposition analyses (SVDAs), between observed and model fields (OM analyses) and amongst the different model runs (MM analyses). Generally, it is found that the first model mode in the MM analysis is very similar to that in the OM analysis, suggesting the model mode with strongest reproducibility is also the mode which tallies best with observations. For the global tropics, the first MM mode is highly reproducible (external variance above 90%) and the first OM couplet is very skillful (correlation between observed and model SVDA time series is over 0.84). The extent to which skill and reproducibility is related to El Nino/Southern Oscillation (ENSO) has been investigated by comparing the OM and MM time series with the Southern Oscillation Index (SOI). For the global tropics, most of the OM and MM variance is common with the SOI, though in boreal summer, the first modes do also have some clear independence from the SOI. The analyses were repeated at the regional scale for Oceania, tropical America, tropical Africa and tropical Southeast Asia. A highly reproducible mode is found in all cases except October-December in Africa. Skill, while always positive, is more variable, strongest for Oceania and tropical America and weakest for Africa. Comparisons with the SOI suggest skill in tropical America and tropical Africa has substantial components that are independent of the SOI, especially in boreal spring and summer when the tropical Atlantic SSTs are strongly related to the leading OM and MM time series. Received: 1 January 1997 / Accepted: 28 July 1997     

南方涛动指数的多元统计分析及其与北半球500hPa月平均环流的关系

本文对美国的CAC南方涛动指数进行了相关分析、因子分析、模糊聚类、主成份分析。指出,月指数可分为二组:1—4月,5—12月,它们的持续性和相关性不相同。还指出,强弱南方涛动可以用涛动指数的第一主成份来定义。 此外,用1951—1984年的月资料讨论了南方涛动指数与当年1月到来年12月的500hPa位势高度场、月平均纬圈高度、月平均纬际高度差、大气活动中心以及大气环流特征量的相互关系。分析表明,在涛动低值年的前期4月份75°N以北的500hPa位势高度明显偏低,5月份偏高;低值年4月第一谐波槽位于东半球并且方差贡献小,5月份一波槽方差贡献大,高值年相反。500hPa位势高度在4—5月的这种变化是涛动在高纬度的一个早期讯号。 最后,我们建立了一个涛动指数的统计预报模式。利用这个模式可成功地预报出1986年的弱涛动并发生El Nino现象。     

SOI和El Nino与湘潭汛期洪涝的关系

本文研究了湘潭汛期（4～9月）洪涝与南方涛动指数（SOI）和厄尔尼诺（ElNino）事件的遥相关联系。结果表明，南方涛动指数异常偏负（弱）以及厄尔尼诺事件的滞后影响对湘潭汛期洪涝的短期气候预报有很好的指示性。在统计分析的基础上，我们建立了日常业务应用的湘潭汛期洪涝预报物理模型。     

纬向平均环流预报的系统性误差及其改进

大量的月预报实例分析表明,纬向平均环流(本指高度场纬向平均分量)存在明显的系统性预报误差,且在总误差中占有可观的份额。国内外其它模式也存在类似的现象。为克服这一困难,本尝试了“结合”(hybrid)的途径。应用重构相空间理论和非线性时空序列预测方法,在大量历史资料的基础上,构造了月尺度逐侯纬向平均高度场(零波分量)距平场的非线性预报模型。然后,将非线性预报和谱模式动力预报结合起来,即将非线性预报结果转化为模式需要的颅报量,再在模式积分过程中的每一步取代其相应部分,实施过程订正。初步试验结果表明,这种途样合效地减少了模式纬向环流的预报误差;特别是通过非线性波流相互作用,还改善了部分波动分量的预报。     

1857—1987年南方涛动指数

本文利用塔希提和达尔文的气压差将美国气候分析中心的SOI序列从1935年向前接补到1857年,并用SOI周期及其与SST的关系进行了检验,证明可以将两段SOI接续起来。     

Low-frequency oscillations in the Northern Hemisphere

Summary Multichannel singular spectrum analysis (MSSA) has been applied to 5-day mean grid point data of 500 hPa geopotential heights over the Northern Hemisphere, to explore low frequency oscillatory variabilities over the region. Two major low-frequency oscillations with periods of 60–70 and 30–40 days were found. These two quasi-periodicities are significantly (at the 99% confidence level) different from those found in the corresponding pure red noise process. Also, they were reproducible when using different window lengths for MSSA and when applying MSSA to two equal-length subsets of the basic data series. In addition, when one subset (first subset) is projected onto the patterns of the other subset, the resulting (projection) spatial-temporal principal components corresponding to the two oscillations are highly correlated with those ST-PCs directly derived from the first subset by MSSA, indicating that the oscillatory components obtained from one subset explain a significant amount of variance in the other subset. Some reliable information about the amplitudes of the oscillations has been also obtained by using composite analysis.With 5 Figures