过去状态估计之集合Kalman滤波扩充状态变量法研究

集合Kalman滤波是由大气数据同化发展的新的同化算法,它利用蒙特卡罗方法计算背景场的误差协方差矩阵,克服了Kalman滤波需要线性化的模型算子和观测算子的难点。但是这种同化方法是一种顺序数据同化方法,无法对过去状态变量进行同化订正。而过去状态的估计对于建立大气或海洋历史资料库、获得准确的数值预报初始场有着重要的意义。本文在集合Kalman滤波同化方法的基础上,提出了可以对过去状态进行估计的集合Kalman滤波扩充状态变量法,然后分别采用空气质量方程和Lorenz系统对这种方法进行了检验。数值试验结果表明,这种方法可以对非线性系统中的过去状态变量进行有效的估计订正,说明该方法是可行的。     

集合KALMAN滤波和最优插值方法在不同观测分布的比较理想试验

目前一种比较流行并且可行的同化方法-集合Kalman滤波(EnKF)能够计算依赖于流的误差统计量。理论上,EnKF能够比最优插值、三维变分等更准确地计算误差统计量,能更好地融合背景场和观测场的信息。作者利用二维平流扩散方程经过10天的同化循环,比较不同观测分布的情况下EnKF和最优插值(OI)的模拟能力。理想试验结果显示,随着观测分布密度的减小,尤其是当观测的分辨率大于OI估计的相关尺度时,集合Kalman滤波的结果比最优插值有更明显的改进。     

T213全球集合预报系统性误差订正研究

针对模式系统性误差一直存在的现状,研究使用卡尔曼滤波的自适应误差订正方法对国家气象中心业务全球集合预报系统的系统性误差进行估计和订正.本文主要介绍这种方法及其原理,其优点是需要的样本量比较小,能够快速经济地对模式产品进行有效的误差估计和订正.使用这种方法对全球T213集合预报系统500 hPa高度场、850 hPa温度场和2 m温度的预报产品进行一阶偏差订正,对订正前后集合预报产品进行检验分析和对比,结果表明,订正后的高层形势场集合预报和2 m温度集合预报的均一性、集合平均的均方根误差和距平相关系数都得到了改善,系统性偏差得到了不同程度的订正,对于存在较大系统性误差的2 m温度预报,订正效果尤其显著.     

全球海洋资料同化系统ZFL_GODAS的研制和初步评估试验

本研究发展了一个全球海洋资料同化系统ZFL_GODAS。该系统是一个短期气候数值预测业务系统的子系统,为短期气候预测海气耦合模式提供全球海洋初始场。系统能够同化的观测资料包括卫星高度计资料、卫星海表温度(SST)资料,以及Argo、XBT、TAO等各种不同来源的现场温盐廓线资料。系统使用的海洋模式为中国科学院大气物理研究所大气科学和地球流体力学数值模拟国家重点实验室开发的气候系统海洋模式LICOM1.0,同化方案为集合最优插值(EnOI)方案。系统使用一个由海洋模式自由积分得到的静态样本来估计背景场误差协方差。这样的基于集合样本的背景场误差协方差具有多变量协变、各向异性的特征,且能反映海洋物理过程固有的空间尺度特征。针对EnOI同化程序的特点,开发了一套特色鲜明、负载均衡、高效的并行化同化程序。本文通过与不同类型观测资料的比较,对同化系统的性能进行了评估。通过比较海表温度和海面高度的年际变率,海表温度异常随时间的变化,SST、海面高度异常(SLA)以及次表层温盐预报产品的均方根误差,5年平均温度偏差廓线、平均盐度廓线、平均纬向流速廓线等发现:系统工作正常、同化效果较好;经过同化以后,各变量都更加接近观测,误差更小,与观测场的相关性更好,可以为短期气候预测系统提供较好的海洋初始场,也可以为物理海洋学的研究提供有效的再分析资料。     

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。     

集合最优插值中的样本选取

背景误差协方差控制了分析向观测调整的幅度以及调整的结构,所以其对同化分析的质量起着至关重要的作用。对于集合同化方法而言,样本决定了背景误差协方差的分布。基于HYCOM海洋数值模式结果,针对集合最优插值方法,探讨了静态样本的选取和更新对背景误差协方差结构分布的影响,研究结果表明：由变量的原始状态数据估计的静态样本会夸大样本相关性,由扣除季节变化得到的距平数据统计出的静态样本能比较合理地反映背景误差协方差的结构;在季风控制区,具有季节变化的样本比静态样本更能适应背景误差协方差随流形分布的特征。同时一系列的同化试验也被实施来进一步调查不同样本对同化分析的影响。     

冷锋影响山东半岛的中尺度客观分析试验

使用山东半岛部分加密测站的锋面个例逐时资料，采用九点拉格朗日插值的逐步订正法，对地面场的资料进行了个尺度客观分析试验。通过对逐步订正法中观测前５个时次资料平均场、观测前１个时次资料场和观测时次资料场作为不同的初始估计场得到的客观分析结果的比较，提出在山东半岛进行中尺度客观分析，当观测资料分布不均，海面区域没有资料点及冷锋附近气象要素变化剧烈时，采用观测时次资料场作为有限区域网格的初始估计场而得到的结果较好。     

Hybrid ETKF-3DVAR方法同化多普勒雷达速度观测资料Ⅰ:模拟资料试验

利用WRF(Weather research and forecasting)模式及模式模拟的资料,采用Hybrid ETKF-3DVAR(ensemble transform Kalman filter-three-dimensional variational data assimilation)方法同化模拟雷达观测资料。该混合同化方法将集合转换卡尔曼滤波(ensemble transform Kalman filter)得到的集合样本扰动通过转换矩阵直接作用到背景场上,利用顺序滤波的思想得到分析扰动场;然后通过增加额外控制变量的方式把"流依赖"的集合协方差信息引入到变分目标函数中去,在3DVAR框架基础下与观测数据进行融合,从而给出分析场的最优估计。试验结果表明,Hybrid ETKF-3DVAR同化方法相比传统3DVAR可以提供更为准确的分析场,Hybrid方法雷达资料初始化模拟的台风涡旋结构与位置比3DVAR更加接近"真实场",对台风路径预报也有明显改进。通过对比Hybrid S试验与Hybrid F试验发现,Hybrid的正效果主要来源于混合背景误差协方差中的"流依赖"信息,集合平均场代替确定性背景场带来的效果并不显著。     

数值预报同化系统中观测资料误差分析

在观测资料同化系统中,观测误差均方差与背景场误差均方差共同决定着观测信息与背景场信息的相对重要性以及这些信息在空间及不同变量间的扩展方式,故在资料同化系统中起到决定性作用.因此,观测误差均方差的合理估计是非常重要的.采用贝塞尔函数拟合方法,从探空观测资料与背景场的偏差(IV)协方差中分离出探空资料的观测误差均方差与模式的背景场误差均方差;再从其他观测资料偏差协方差中扣除背景场误差均方差来估计其它观测误差均方差.观测误差均方差分析使用2006年8月1-31日观测资料,国家气象中心T213L31全球中期分析预报系统的6小时预报作为背景场.结果显示估计的观测误差均方差是比较合理的.     

基于Hybrid EnSRF-En3DVar的雷达资料同化研究

基于WRF模式构建了Hybrid En SRF-En3DVar同化系统,该系统使用En SRF方案直接更新集合扰动。利用构建的同化系统针对台风"桑美"分别进行集合协方差权重敏感性试验和同化雷达不同观测资料的敏感性试验。集合协方差权重敏感性试验发现:当集合协方差权重分别为0.25、0.5和0.75时,同化效果优于3DVar试验,其中0.75的集合协方差权重试验得到了分析场的最优估计;当集合协方差权重为1.0时,分析场最差。同化雷达不同观测资料的敏感性试验表明,联合同化雷达径向风及反射率能有效改善大气湿度场和风场,但对风场的改善效果不如仅同化雷达径向风好。将En SRF集合扰动更新方案与扰动观测方案综合分析发现,扰动观测方案集合离散度较小,计算代价大,En SRF方案优于扰动观测方案。     

A study on air–sea interaction on the simulated seasonal climate in an ocean–atmosphere coupled model

This study investigates the effects of air–sea interaction upon simulated tropical climatology, focusing on the boreal summer mean precipitation and the embedded intra-seasonal oscillation (ISO) signal. Both the daily coupling of ocean–atmosphere and the diurnal variation of sea surface temperature (SST) at every time step by accounting for the ocean mixed layer and surface-energy budget at the ocean surface are considered. The ocean–atmosphere coupled model component of the global/regional integrated model system has been utilized. Results from the coupled model show better precipitation climatology than those from the atmosphere-only model, through the inclusion of SST–cloudiness–precipitation feedback in the coupled system. Cooling the ocean surface in the coupled model is mainly responsible for the improved precipitation climatology, whereas neither the coupling itself nor the diurnal variation in the SST influences the simulated climatology. However, the inclusion of the diurnal cycle in the SST shows a distinct improvement of the simulated ISO signal, by either decreasing or increasing the magnitude of spectral powers, as compared to the simulation results that exclude the diurnal variation of the SST in coupled models.     

The influence of climate regime shift on ENSO

Observations indicated that for the El Niño/Southern Oscillation (ENSO) there have been eastward displacements of the zonal wind stress (WS) anomalies and surface heat flux (short wave heat flux and latent heat flux) anomalies during El Niño episodes in the 1981–1995 regime relative to the earlier regime of 1961–1975 (without corresponding displacements during La Niña episodes). Our numerical experiments with the Zebiak–Cane coupled model generally reproduced such displacements when the model climatological fields were replaced by the observed climatologies [of sea surface temperature (SST), surface WS and surface wind atmospheric divergence] and simulated climatologies (of oceanic surface layer currents and associated upwelling) for the 1981–1995 regime. Sensitivity tests indicated that the background atmospheric state played a much more important role than the background ocean state in producing the displacements, which enhanced the asymmetry between El Niño and La Niña in the later regime. The later regime climatology state resulted in the eastward shifts in the ENSO system (WS and SST) only during El Niño, through the eastward shift of the atmosphere convergence heating rate in the coupled model. The ENSO period and ENSO predictability were also enhanced in the coupled model under the later regime climatology. That the change in the mean state of the tropical Pacific atmosphere and ocean after the mid 1970s could have produced the observed changes in ENSO properties is consistent with our findings.     

Simulation of salinity variability and the related freshwater flux forcing in the tropical Pacific: An evaluation using the Beijing normal university earth system model (BNU-ESM)

The climatology and interannual variability of sea surface salinity(SSS) and freshwater flux(FWF) in the equatorial Pacific are analyzed and evaluated using simulations from the Beijing Normal University Earth System Model(BNU-ESM).The simulated annual climatology and interannual variations of SSS, FWF, mixed layer depth(MLD), and buoyancy flux agree with those observed in the equatorial Pacific. The relationships among the interannual anomaly fields simulated by BNU-ESM are analyzed to illustrate the climate feedbacks induced by FWF in the tropical Pacific. The largest interannual variations of SSS and FWF are located in the western-central equatorial Pacific. A positive FWF feedback effect on sea surface temperature(SST) in the equatorial Pacific is identified. As a response to El Ni ?no–Southern Oscillation(ENSO),the interannual variation of FWF induces ocean processes which, in turn, enhance ENSO. During El Ni ?no, a positive FWF anomaly in the western-central Pacific(an indication of increased precipitation rates) acts to enhance a negative salinity anomaly and a negative surface ocean density anomaly, leading to stable stratification in the upper ocean. Hence, the vertical mixing and entrainment of subsurface water into the mixed layer are reduced, and the associated El Ni ?no is enhanced. Related to this positive feedback, the simulated FWF bias is clearly reflected in SSS and SST simulations, with a positive FWF perturbation into the ocean corresponding to a low SSS and a small surface ocean density in the western-central equatorial Pacific warm pool.     

Impacts of assimilated data on reanalysis climatology

Impacts of the observing systems and meteorological parameters on a reanalyzed climatology are investigated using an assimilation algorithm based on the National Centers for Environmental Prediction/Department of Energy (NCEP/DOE) reanalysis system. The one-year analyzed climatology is generated in a standard analysis using all of the observations, which is named the control run. Additional climatology data sets are produced by selectively choosing observational systems or meteorological variables in the reanalysis system. It is found that the radiosonde-only observation is sufficient for reproducing the reanalysis climatology. Satellite observations have no significant contribution to the large-scale fields. The importance of meteorological variables on the analyzed climatology is temperature, wind, and moisture in that order, where the mass observation is found to have a greater impact on the analyzed climatology than wind. It is, however, noted that the moisture field demonstrates a profound influence on the surface hydroclimate such as cloud cover, radiation flux, and land surface temperature.     

The ACPI Project, Element 1: Initializing a Coupled Climate Model from Observed Conditions

A problem for climate change studies with coupled ocean-atmosphere models has been how to incorporate observed initial conditions into the ocean, which holds most of the `memory' of anthropogenic forcing effects. The first difficulty is the lack of comprehensive three-dimensional observations of the current ocean temperature (T) and salinity (S) fields to initialize to. The second problem is that directly imposing observed T and S fields into the model results in rapid drift back to the model climatology, with the corresponding loss of the observed information. Anthropogenic forcing scenarios therefore typically initialize future runs by starting with pre-industrial conditions. However, if the future climate depends on the details of the present climate, then initializing the model to observations may provide more accurate forecasts. Also, this ～ 130 yr spin up imposes substantial overhead if only a few decades of predictions are desired. A new technique to address these problems is presented. In lieu of observed T and S, assimilated ocean data were used. To reduce model drift, an anomaly coupling scheme was devised. This consists of letting the model's climatological (pre-industrial) oceanic and atmospheric heat contents and transports balance each other, while adding on the (much smaller) changes in heat content since the pre-industrial era as anomalies. The result is model drift of no more than 0.2 K over 50 years, significantly smaller than the forced response of 1.0 K. An ensemble of runs with these assimilated initial conditions is then compared to a set spun up from pre-industrial conditions. No systematic differences were found, i.e., the model simulation of the ocean temperature structure in the late 1990s is statistically indistinguishable from the assimilated observations. However, a model with a worse representation of the late 20th century climate might show significant differences if initialized in this way.     

A Regional Ocean Reanalysis System for Coastal Waters of China and Adjacent Seas

A regional ocean reanalysis system for the coastal waters of China and adjacent seas has been developed by the National Marine Data and Information Service(NMDIS).It produces a dataset package called CORA (China ocean reanalysis).The regional ocean model used is based on the Princeton Ocean Model with a generalized coordinate system(POMgcs).The model is parallelized by NMDIS with the addition of the wave breaking and tidal mixing processes into model parameterizations.Data assimilation is a sequential three-dimensional variational(3D-Var) scheme implemented within a multigrid framework.Observations include satellite remote sensing sea surface temperature(SST),altimetry sea level anomaly(SLA),and temperature/salinity profiles.The reanalysis fields of sea surface height,temperature,salinity,and currents begin with January 1986 and are currently updated every year. Error statistics and error distributions of temperature,salinity and currents are presented as a primary evaluation of the reanalysis fields using sea level data from tidal gauges,temperature profiles,as well as the trajectories of Argo floats.Some case studies offer the opportunity to verify the evolution of certain local circulations.These evaluations show that the reanalysis data produced provide a good representation of the ocean processes and phenomena in the coastal waters of China and adjacent seas.     

Importance of Air-Sea Coupling in Simulating Tropical Cyclone Intensity at Landfall

An atmosphere-only model system for making seasonal prediction and projecting future intensities of landfalling tropical cyclones (TCs) along the South China coast is upgraded by including ocean and wave models. A total of 642 TCs have been re-simulated using the new system to produce a climatology of TC intensity in the South China Sea. Detailed comparisons of the simulations from the atmosphere-only and the fully coupled systems reveal that the inclusion of the additional ocean and wave models enable differential sea surface temperature responses to various TC characteristics such as translational speed and size. In particular, interaction with the ocean does not necessarily imply a weakening of the TC, with the coastal bathymetry possibly playing a role in causing a near-shore intensification of the TC. These results suggest that to simulate the evolution of TC structure more accurately, it is essential to use an air-sea coupled model instead of an atmosphere-only model.     

The UVic earth system climate model: Model description,climatology, and applications to past,present and future climates

Abstract

A new earth system climate model of intermediate complexity has been developed and its climatology compared to observations. The UVic Earth System Climate Model consists of a three‐dimensional ocean general circulation model coupled to a thermodynamic/dynamic sea‐ice model, an energy‐moisture balance atmospheric model with dynamical feedbacks, and a thermomechanical land‐ice model. In order to keep the model computationally efficient a reduced complexity atmosphere model is used. Atmospheric heat and freshwater transports are parametrized through Fickian diffusion, and precipitation is assumed to occur when the relative humidity is greater than 85%. Moisture transport can also be accomplished through advection if desired. Precipitation over land is assumed to return instantaneously to the ocean via one of 33 observed river drainage basins. Ice and snow albedo feedbacks are included in the coupled model by locally increasing the prescribed latitudinal profile of the planetary albedo. The atmospheric model includes a parametrization of water vapour/planetary longwave feedbacks, although the radiative forcing associated with changes in atmospheric CO₂ is prescribed as a modification of the planetary longwave radiative flux. A specified lapse rate is used to reduce the surface temperature over land where there is topography. The model uses prescribed present‐day winds in its climatology, although a dynamical wind feedback is included which exploits a latitudinally‐varying empirical relationship between atmospheric surface temperature and density. The ocean component of the coupled model is based on the Geophysical Fluid Dynamics Laboratory (GFDL) Modular Ocean Model 2.2, with a global resolution of 3.6° (zonal) by 1.8° (meridional) and 19 vertical levels, and includes an option for brine‐rejection parametrization. The sea‐ice component incorporates an elastic‐viscous‐plastic rheology to represent sea‐ice dynamics and various options for the representation of sea‐ice thermodynamics and thickness distribution. The systematic comparison of the coupled model with observations reveals good agreement, especially when moisture transport is accomplished through advection.

Global warming simulations conducted using the model to explore the role of moisture advection reveal a climate sensitivity of 3.0°C for a doubling of CO₂, in line with other more comprehensive coupled models. Moisture advection, together with the wind feedback, leads to a transient simulation in which the meridional overturning in the North Atlantic initially weakens, but is eventually re‐established to its initial strength once the radiative forcing is held fixed, as found in many coupled atmosphere General Circulation Models (GCMs). This is in contrast to experiments in which moisture transport is accomplished through diffusion whereby the overturning is reestablished to a strength that is greater than its initial condition.

When applied to the climate of the Last Glacial Maximum (LGM), the model obtains tropical cooling (30°N‐30°S), relative to the present, of about 2.1°C over the ocean and 3.6°C over the land. These are generally cooler than CLIMAP estimates, but not as cool as some other reconstructions. This moderate cooling is consistent with alkenone reconstructions and a low to medium climate sensitivity to perturbations in radiative forcing. An amplification of the cooling occurs in the North Atlantic due to the weakening of North Atlantic Deep Water formation. Concurrent with this weakening is a shallowing of, and a more northward penetration of, Antarctic Bottom Water.

Climate models are usually evaluated by spinning them up under perpetual present‐day forcing and comparing the model results with present‐day observations. Implicit in this approach is the assumption that the present‐day observations are in equilibrium with the present‐day radiative forcing. The comparison of a long transient integration (starting at 6 KBP), forced by changing radiative forcing (solar, CO₂, orbital), with an equilibrium integration reveals substantial differences. Relative to the climatology from the present‐day equilibrium integration, the global mean surface air and sea surface temperatures (SSTs) are 0.74°C and 0.55°C colder, respectively. Deep ocean temperatures are substantially cooler and southern hemisphere sea‐ice cover is 22% greater, although the North Atlantic conveyor remains remarkably stable in all cases. The differences are due to the long timescale memory of the deep ocean to climatic conditions which prevailed throughout the late Holocene. It is also demonstrated that a global warming simulation that starts from an equilibrium present‐day climate (cold start) underestimates the global temperature increase at 2100 by 13% when compared to a transient simulation, under historical solar, CO₂ and orbital forcing, that is also extended out to 2100. This is larger (13% compared to 9.8%) than the difference from an analogous transient experiment which does not include historical changes in solar forcing. These results suggest that those groups that do not account for solar forcing changes over the twentieth century may slightly underestimate (～3% in our model) the projected warming by the year 2100.     

我国逐日降水量格点化方法

国家气象信息中心(NMIC)和美国大气海洋局气候预测中心合作开发了"中国逐日格点降水量实时分析系统(V1.0)",并已在NMIC投入业务试运行。该系统基于我国2419个国家级地面气象站日降水量观测(08:00—08:00,北京时)数据,采用"基于气候背景场"的最优插值方法,实时生成空间分辨率为0.5°×0.5°的格点化日降水量资料。通过对汛期典型区域和单站降水过程的对比分析表明:该格点化产品的精度较高,能准确捕捉并再现每一次降水过程。误差分析表明:约91%的数据绝对误差小于1.0 mm/d。该产品在定量分析天气实况、检验天气气候模式精度、检验卫星产品精度等方面有应用前景。