Coupling Ensemble Kalman Filter with Four-dimensional Variational Data Assimilation

This study examines the performance of coupling the deterministic four-dimensional variational assimilation system (4DVAR) with an ensemble Kalman filter (EnKF) to produce a superior hybrid approach for data assimilation. The coupled assimilation scheme (E4DVAR) benefits from using the state-dependent uncertainty provided by EnKF while taking advantage of 4DVAR in preventing filter divergence: the 4DVAR analysis produces posterior maximum likelihood solutions through minimization of a cost function about which the ensemble perturbations are transformed, and the resulting ensemble analysis can be propagated forward both for the next assimilation cycle and as a basis for ensemble forecasting. The feasibility and effectiveness of this coupled approach are demonstrated in an idealized model with simulated observations. It is found that the E4DVAR is capable of outperforming both 4DVAR and the EnKF under both perfect- and imperfect-model scenarios. The performance of the coupled scheme is also less sensitive to either the ensemble size or the assimilation window length than those for standard EnKF or 4DVAR implementations.     

集合Kalman滤波资料同化技术及研究现状

针对国内集合Kalman滤波资料同化领域的研究空白,对该技术的背景、理论、优势以及存在的问题做了简要描述,对目前国际上的主要研究成果做了介绍,并给出了该方法可能的发展方向。     

Nonlinear measurement function in the ensemble Kalman filter

ABSTRACT The optimal Kalman gain was analyzed in a rigorous statistical framework. Emphasis was placed on a comprehensive understanding and interpretation of the current algorithm, especially when the measurement function is nonlinear. It is argued that when the measurement function is nonlinear, the current ensemble Kalman Filter algorithm seems to contain implicit assumptions： the forecast of the measurement function is unbiased or the nonlinear measurement function is linearized. While the forecast of the model state is assumed to be unbiased, the two assumptions are actually equivalent. On the above basis, we present two modified Kalman gain algorithms. Compared to the current Kalman gain algorithm, the modified ones remove the above assumptions, thereby leading to smaller estimated errors. This outcome was confirmed experimentally, in which we used the simple Lorenz 3-component model as the test-bed. It was found that in such a simple nonlinear dynamical system, the modified Kalman gain can perform better than the current one. However, the application of the modified schemes to realistic models involving nonlinear measurement functions needs to be further investigated.     

集合卡尔曼滤波数据同化在一维波动方程中的应用

简要回顾了集合卡尔曼滤波(EnKF：Ensemble Kalman Filter)数据同化方法的发展历史,并介绍了EnKF数据同化方法的基本原理,利用一维非线性波动方程进行了数值试验。EnKF数据同化方法的实现过程简单可行。避免了EKF中协方差演变方程预报过程中出现的计算不准确和关于协方差矩阵的大量数据的存储问题,最主要的是EnKF可以有效控制模式变量估计误差方差的增长,改善预报效果。     

集合Kalman滤波在土壤湿度同化中的应用

基于非饱和土壤水模型和集合卡尔曼滤波 (Ensemble Kalman Filter, 简称EnKF) 并结合陆面水文模型——可变下渗能力模型 (Variable Infiltration Capacity, 简称VIC模型) 发展了一个土壤湿度同化方案。利用1998年6~8月淮河流域能量和水循环试验 (HUBEX) 项目外场观测试验区——史灌河流域梅山站土壤湿度逐日观测资料及1986~1993年合肥和南阳两站点的土壤湿度旬观测资料进行同化试验, 结果表明该同化方案能完整估计土壤湿度廓线, 同化的土壤湿度与观测资料基本吻合, 反映了土壤湿度的日、 旬、 月、 季变化, 同化方案是合理的。与基于扩展卡尔曼滤波 (Extended Kalman Filter, 简称EKF) 的土壤湿度同化方案的结果比较, 基于EnKF的土壤湿度同化方案易于实现, 且通过选择恰当的集合样本数其同化效果总体上略优于EKF同化方案, 但前者同化时需要花费较多的计算时间。     

An Adaptive Estimation of Forecast Error Covariance Parameters for Kalman Filtering Data Assimilation

An adaptive estimation of forecast error covariance matrices is proposed for Kalman filtering data assimilation. A forecast error covariance matrix is initially estimated using an ensemble of perturbation forecasts. This initially estimated matrix is then adjusted with scale parameters that are adaptively estimated by minimizing -2log-likelihood of observed-minus-forecast residuals. The proposed approach could be applied to Kalman filtering data assimilation with imperfect models when the model error statistics are not known. A simple nonlinear model (Burgers' equation model) is used to demonstrate the efficacy of the proposed approach.     

不同模式误差方案在集合Kalman滤波土壤湿度同化中的比较试验

本文主要目的是探讨不同模式误差方案在土壤湿度同化中的性能。基于集合Kalman滤波同化方法和AVIM (Atmosphere-Vegetation Interaction Model) 陆面模式, 利用理想试验对膨胀因子方案 (Covariance Inflation, 简称CI)、 直接随机扰动方案 (Direct Random Disturbance, 简称DRD)、 误差源扰动方案 (Source Random Disturbance, 简称SRD) 等3种模式误差方案的同化效果进行了比较, 讨论了各方案在不同观测误差、 观测层数、 观测间隔情况下的同化性能。试验结果表明在观测误差估计完全准确的情况下, 3种方案都能获得较好的同化效果, 并且SRD方案相对于真值的均方根误差最小。当观测误差估计不准确时, SRD方案的同化效果仍能基本得以保持, 而CI和DRD方案则对观测误差估计更为敏感, 同化效果下降明显。当同化多层观测时, CI和DRD方案由于难以保持不同层观测之间的匹配关系, 同化结果反而变差, 而SRD方案能有效协调同化多层观测, 增加观测层后同化结果有了进一步的改善。当观测时间间隔较大时, CI和DRD方案的同化效果显著下降; 而SRD方案由于包含了一定的误差订正功能, 在观测稀疏时仍能保持较好的同化效果。     

UKF滤波算法在数据同化中的应用

介绍了一种新的数据同化算法（UKF,Unscented Kalman Filter）,该算法不需要计算伴随矩阵,就能够解决模式的非线性问题。以Lorenz系统为例,进行了数据同化的数值试验。结果表明：基于UKF的同化方案与背景场的初始值无关,它能有效地抑制状态变量误差的增长,同化结果精度高。     

Data Assimilation Method Based on the Constraints of Confidence Region

The ensemble Kalman filter(En KF) is a distinguished data assimilation method that is widely used and studied in various fields including methodology and oceanography. However, due to the limited sample size or imprecise dynamics model, it is usually easy for the forecast error variance to be underestimated, which further leads to the phenomenon of filter divergence.Additionally, the assimilation results of the initial stage are poor if the initial condition settings differ greatly from the true initial state. To address these problems, the variance inflation procedure is usually adopted. In this paper, we propose a new method based on the constraints of a confidence region constructed by the observations, called En CR, to estimate the inflation parameter of the forecast error variance of the En KF method. In the new method, the state estimate is more robust to both the inaccurate forecast models and initial condition settings. The new method is compared with other adaptive data assimilation methods in the Lorenz-63 and Lorenz-96 models under various model parameter settings. The simulation results show that the new method performs better than the competing methods.     

集合卡尔曼平滑和集合卡尔曼滤波在污染源反演中的应用

此文目的是讨论污染源反演问题的统计方法.基于Bayes估计理论,该文将资料同化中的集合平滑、集合卡尔曼平滑和集合卡尔曼滤波应用在污染源反演问题中.在详细给出污染源反演的集合平滑、集合卡尔曼平滑和集合卡尔曼滤波的严格数学表达后,用一个简单的模型演示了集合卡尔曼平滑和集合卡尔曼滤波在污染源反演中的可行性,并且通过对比理想试验结果比较了集合卡尔曼平滑和集合卡尔曼滤波方法在反演污染源排放的效果,讨论了观测误差和污染源先验误差估计对反演结果的影响.试验结果表明在观测间隔小和观测误差小的情况下,集合卡尔曼滤波和集合卡尔曼平滑都可以有效地反演出随时间变化的污染源排放.当观测误差增大时,集合卡尔曼滤波和集合卡尔曼平滑的反演效果都有一定降低,但是反演误差的增加少于观测误差的增加,同时集合卡尔曼平滑(Ensemble Kalman smoother,简称EnKS)对观测误差比集合卡尔曼滤波(Ensemble Kalman filter,简称EnKF)更为敏感.当观测时间间隔较大时,EnKF不能对没有观测时的污染源排放进行估计,仅能对有观测时的污染源排放进行较好的反演.而EnKS可以利用观测对观测时刻前的污染源排放进行反演,因此其效果明显好于EnKF,并且在观测时间间隔较大的情况下依然可以较好地反演出污染源排放.试验结果还显示污染源排放的先验误差估计对反演的结果有较大影响.     

Assimilating Best Track Minimum Sea Level Pressure Data Together with Doppler Radar Data Using an Ensemble Kalman Filter for Hurricane Ike (2008) at a Cloud-Resolving Resolution

Extending an earlier study, the best track minimum sea level pressure (MSLP) data are assimilated for landfalling Hurricane Ike (2008) using an ensemble Kalman filter (EnKF), in addition to data from two coastal ground-based Doppler radars, at a 4-km grid spacing. Treated as a sea level pressure observation, the MSLP assimilation by the EnKF enhances the hurricane warm core structure and results in a stronger and deeper analyzed vortex than that in the GFS (Global Forecast System) analysis; it also improves the subsequent 18-h hurricane intensity and track forecasts. With a 2-h total assimilation window length, the assimilation of MSLP data interpolated to 10-min intervals results in more balanced analyses with smaller subsequent forecast error growth and better intensity and track forecasts than when the data are assimilated every 60 minutes. Radar data are always assimilated at 10-min intervals. For both intensity and track forecasts, assimilating MSLP only outperforms assimilating radar reflectivity (Z) only. For intensity forecast, assimilating MSLP at 10-min intervals outperforms radar radial wind (Vr) data (assimilated at 10-min intervals), but assimilating MSLP at 60-min intervals fails to beat Vr data. For track forecast, MSLP assimilation has a slightly (noticeably) larger positive impact than Vr(Z) data. When Vr or Z is combined with MSLP, both intensity and track forecasts are improved more than the assimilation of individual observation type. When the total assimilation window length is reduced to 1h or less, the assimilation of MSLP alone even at 10-min intervals produces poorer 18-h intensity forecasts than assimilating Vr only, indicating that many assimilation cycles are needed to establish balanced analyses when MSLP data alone are assimilated; this is due to the very limited pieces of information that MSLP data provide.     

Comparison of nonlinear local Lyapunov vectors with bred vectors,random perturbations and ensemble transform Kalman filter strategies in a barotropic model

The breeding method has been widely used to generate ensemble perturbations in ensemble forecasting due to its simple concept and low computational cost. This method produces the fastest growing perturbation modes to catch the growing components in analysis errors. However, the bred vectors(BVs) are evolved on the same dynamical flow, which may increase the dependence of perturbations. In contrast, the nonlinear local Lyapunov vector(NLLV) scheme generates flow-dependent perturbations as in the breeding method, but regularly conducts the Gram–Schmidt reorthonormalization processes on the perturbations. The resulting NLLVs span the fast-growing perturbation subspace efficiently, and thus may grasp more components in analysis errors than the BVs.In this paper, the NLLVs are employed to generate initial ensemble perturbations in a barotropic quasi-geostrophic model.The performances of the ensemble forecasts of the NLLV method are systematically compared to those of the random perturbation(RP) technique, and the BV method, as well as its improved version—the ensemble transform Kalman filter(ETKF)method. The results demonstrate that the RP technique has the worst performance in ensemble forecasts, which indicates the importance of a flow-dependent initialization scheme. The ensemble perturbation subspaces of the NLLV and ETKF methods are preliminarily shown to catch similar components of analysis errors, which exceed that of the BVs. However, the NLLV scheme demonstrates slightly higher ensemble forecast skill than the ETKF scheme. In addition, the NLLV scheme involves a significantly simpler algorithm and less computation time than the ETKF method, and both demonstrate better ensemble forecast skill than the BV scheme.     

Assimilation of Hourly Surface Observations with the Canadian High-Resolution Ensemble Kalman Filter

An hourly-cycling ensemble Kalman filter (EnKF) working at 2.5?km horizontal grid spacing is implemented over southern Ontario (Canada) to assimilate Meteorological Terminal Aviation Routine Weather Reports (METARs) in addition to the observations assimilated operationally at the Canadian Meteorological Centre. This high-resolution EnKF (HREnKF) system employs ensemble land analyses and perturbed roughness length to prevent an ensemble spread that is too small near the surface. The HREnKF then performs continuously for a four-day period, from which twelve-hour ensemble forecasts are launched every six hours. The impact on analyses and short-term forecasts of assimilating METAR data is given special attention.

It is shown that using ensemble land surface analyses increases near-surface ensemble spreads for temperature and specific humidity. Perturbing roughness length enlarges the spread for surface wind. Given sufficient ensemble spread, the four-day case study shows that the near-surface model state is brought closer to surface observations during the cycling process. The impact of assimilating surface data can also be seen at higher levels by using aircraft reports for verification. The ensemble forecast verification suggests that METAR data assimilation improves ensemble forecasts of air temperature and dewpoint near the surface up to a lead time of six hours or even longer. However, only minor improvement is found in surface wind forecasts.     

集合卡尔曼滤波同化多普勒雷达资料的数值试验

利用集合卡尔曼滤波(EnKF)在云数值模式中同化模拟多普勒雷达资料,并考察了不同条件下EnKF同化方法的性能.结果显示,经过几个同化周期后,EnKF分析结果非常接近真值.单多普勒雷达资料EnKF同化对雷达位置不太敏感,双雷达资料同化结果在同化的初期阶段比单雷达资料同化结果准确.同化由反射率导出的雨水比直接同化反射率资料更有效,联合同化径向速度和雨水有利于提高同化分析效果.协方差对EnKF同化效果起着非常重要的作用,考虑模式全部预报变量与径向速度协方差的同化效果比仅考虑速度场与径向速度协方差的同化效果好.雷达资料缺值降低了同化效果,此时增加地面常规观测资料的同化可以明显提高同化分析效果.EnKF同化技术对雷达观测资料误差不太敏感.初始集合对同化分析有较大影响.EnKF同化受集合大小和观测资料影响半径.同化对模式误差较敏感.利用EnKF同化双多普勒雷达资料,分析了一次梅雨锋暴雨过程的中尺度结构.结果表明,EnKF同化技术能够从双多普勒雷达资料反演暴雨中尺度系统的动力场、热力场和微物理场,反演的风场是较准确的,反演的热力场和微物理场分布也是基本合理的.中低层切变线是此次暴雨的主要动力特征,对流云表现为低层辐合、高层辐散并有垂直上升运动伴随,其热力特征表现为低层是低压区,高层为高压区,中部为暖区而上、下部为冷区,水汽、云水和雨水分别集中在对流云体内、上升气流区和强回波区.     

关于集合Kalman滤波的理论和方法的发展

随着同化方法的不断的发展，一种新的资料同化方法——集合Kalman滤波正在兴起。简单地回顾了同化方法的发展，探讨了集合Kalman滤波的特点。同时，还介绍了集合Kalman滤波发展的过程以及指出目前所面临的问题和未来的发展趋势。     

Using analysis state to construct a forecast error covariance matrix in ensemble Kalman filter assimilation

Correctly estimating the forecast error covariance matrix is a key step in any data assimilation scheme. If it is not correctly estimated, the assimilated states could be far from the true states. A popular method to address this problem is error covariance matrix inflation. That is, to multiply the forecast error covariance matrix by an appropriate factor. In this paper, analysis states are used to construct the forecast error covariance matrix and an adaptive estimation procedure associated with the error covariance matrix inflation technique is developed. The proposed assimilation scheme was tested on the Lorenz-96 model and 2D Shallow Water Equation model, both of which are associated with spatially correlated observational systems. The experiments showed that by introducing the proposed structure of the forecast error covariance matrix and applying its adaptive estimation procedure, the assimilation results were further improved.     

臭氧卫星观测资料同化系统构建及其试验研究

为检验臭氧卫星资料同化对臭氧分析场和预报场的影响,基于集合平方根滤波(ENSRF)理论,结合通用地球系统模式(CESM),构建了CESM-ENSRF同化预报系统。系统构建过程考虑了卡尔曼滤波同化中的关键问题:利用全场随机扰动对初始场加扰,结合一般协方差膨胀和松弛协方差膨胀方法实现协方差膨胀,使用五阶距离相关函数进行协方差局地化。将构建的系统用于微波临边探测器(MLS)臭氧廓线数据的同化,分析臭氧卫星资料同化对模式预报的影响。结果表明:构建的CESM-ENSRF同化系统有效实现了臭氧资料同化,臭氧卫星资料同化对臭氧分析场和预报场精度有较大改进。     

基于集合Kalman滤波数据同化的热带气旋路径集合预报研究

构建了一个基于集合Kalman滤波数据同化的热带气旋集合预报系统,通过积云参数化方案和边界层参数化方案的9个不同组合,采用MM5模式进行了不同时间的短时预报。对预报结果使用“镜像法”得到18个初始成员,为同化提供初始背景集合。将人造台风作为观测场,同化后的结果作为集合预报的初值,通过不同参数组合的MM5模式进行集合预报。对2003～2004年16个台风个例的分析表明,初始成员产生方法能够对热带气旋的要素场、中心强度和位置进行合理扰动。同化结果使台风强度得到加强,结构更接近实际。基于同化的集合路径预报结果要优于未同化的集合预报。使用“镜像法”增加集合成员提高了预报准确度,路径预报误差在48小时和72小时分别低于200 km和250 km。     

A Multimodel Ensemble-based Kalman Filter for the Retrieval of Soil Moisture Profiles

With the combination of three land surface models (LSMs) and the ensemble Kalman filter (EnKF), a multimodel EnKF is proposed in which the multimodel background superensemble error covariance matrix is estimated by two different algorithms: the Simple Model Average (SMA) and the Weighted Average Method (WAM). The two algorithms are tested and compared in terms of their abilities to retrieve the true soil moisture profile by respectively assimilating both synthetically-generated and actual near-surface soil moisture measurements. The results from the synthetic experiment show that the performances of the SMA and WAM algorithms were quite different. The SMA algorithm did not help to improve the estimates of soil moisture at the deep layers, although its performance was not the worst when compared with the results from the single-model EnKF. On the contrary, the results from the WAM algorithm were better than those from any single-model EnKF. The tested results from assimilating the field measurements show that the performance of the two multimodel EnKF algorithms was very stable compared with the single-model EnKF. Although comparisons could only be made at three shallow layers, on average, the performance of the WAM algorithm was still slightly better than that of the SMA algorithm. As a result, the WAM algorithm should be adopted to approximate the multimodel background superensemble error covariance and hence used to estimate soil moisture states at the relatively deep layers.     

WRF-EnSRF卫星资料同化模块的构建研究

为加强国内卫星资料在同化系统中的应用,在自主构建的新一代WRF-EnSRF同化系统中,采用RTTOV辐射传输模式作为观测算子,并建立卫星资料读取、偏差订正及质量控制等子模块,构建出WRF-EnSRF卫星资料同化系统.运用该同化系统,同时同化NOAA-16的AMSU-A和AMSU-B的辐射率资料,进行华南暴雨过程的卫星资料同化数值模拟试验.试验结果表明:偏差订正后亮温资料拟合结果基本位于主对角线上,偏差有所降低.从TS评分看,同化试验对中雨及大雨部分的降水落区以及暴雨级别以上的降水强度的模拟效果有改善.试验证明,建立的卫星同化系统是可运行的.