Pressure image assimilation for atmospheric motion estimation

The complexity of the laws of dynamics governing 3-D atmospheric flows associated with incomplete and noisy observations make the recovery of atmospheric dynamics from satellite image sequences very difficult. In this paper, we address the challenging problem of estimating physical sound and time-consistent horizontal motion fields at various atmospheric depths for a whole image sequence. Based on a vertical decomposition of the atmosphere, we propose a dynamically consistent atmospheric motion estimator relying on a multilayer dynamic model. This estimator is based on a weak constraint variational data assimilation scheme and is applied on noisy and incomplete pressure difference observations derived from satellite images. The dynamic model is a simplified vorticity-divergence form of a multilayer shallow-water model. Average horizontal motion fields are estimated for each layer. The performance of the proposed technique is assessed using synthetic examples and using real world meteorological satellite image sequences. In particular, it is shown that the estimator enables exploiting fine spatio-temporal image structures and succeeds in characterizing motion at small spatial scales.     

A 4D-variational approach applied to an eddy-permitting North Atlantic configuration: Synthetic and real data assimilation of altimeter observations

The increasing number of oceanic observations calls for the use of synthetic methods to provide consistent analyses of the oceanic variability that will support a better understanding of the underlying mechanisms. In this study, a 1/3° eddy-permitting model of the North Atlantic (from 20°S to 70°N) is combined with a 4D-variational method to estimate the oceanic state from altimeter observations. This resolution allows a better extraction of the physical content of altimeter data since the model spatial scales are more consistent with the data than coarser assimilation exercises because of a lower error in model representativity. Several strategies for the assimilation window are tested through twin experiments carried out under the following conditions: different window lengths and either a quasi-static (also known as progressive) variational assimilation with progressive extension of the window, or a simpler direct method without prior assimilation. From our set of experiments, the most efficient strategy is the use of both a simple direct assimilation method and a 90-day window. The assimilation of synthetic altimeter data constrains the model-temperature, -salinity and -velocity fields mainly over the first 1300 m where the error is the largest. Improvements occur not only in quiescent regions, but also in more energetic meso-scale regimes. Despite the existence of model- and surface forcing-errors as well as large errors in the first guess, the assimilation of real altimeter data proves to be consistent with our twin experiments. Indeed, the analyses show a better detachment of the Gulf Stream, weaker regional biases and more accurate positions for meso-scale structures. Independent hydrographic data (Argo floats and CTD cruises) and transports estimates along the OVIDE 2002 cruise show an improvement of the analysed oceanic state with respect to the assimilation-free case though water mass properties are still incorrectly represented. After assimilation, the North Atlantic heat transport in the model is in good agreement with independent estimates based on hydrographic data.     

Numerical experiments with data assimilation in the simplest dynamic-stochastic model for the ocean's synoptic variability closed at the level of the second moments

The results of numerical experiments with data assimilation in a non-linear model of the synoptic dynamics of the world's oceans are reported. The model's response to the data assimilation is studied at different values of the relative error in determining the initial fields and assimilation discreteness. The results of application of the algorithms of optimum filtration and optimum interpolation are compared. The advantages of the filtration algorithm employed are discussed only for the case of small errors in determining the initial fields and low assimilation discreteness. The possibility of deterioration of the prediction estimates in data assimilation with discreteness less than the typical time of variability of the average fields is noted.Translated by Mikhail M. Trufanov.     

Physical and biological modeling in the Gulf Stream region: Part II. Physical and biological processes

Mesoscale physical and biological processes are examined at the Gulf Stream front by means of a 4-D simulation including physical and biological data assimilation. The data assimilated are from Leg 1 of the Fall BIOSYNOP cruise, 21 Sept.–8 Oct. 1988, and GULFCAST data for the same period. Focus is on the vertical velocities at the front, the vertical and horizontal transports of nutrients and plankton, and the impact of these transports on phytoplankton biomass, production and organic particle export. It was found that while jet meandering enhances new production at the front, primary production and phytoplankton concentration at the front are not significantly enhanced over those of Slope water. Winds during this period also have little impact on productivity at the front, due to their high temporal variability. Ring–stream interactions, however, significantly increase the net vertical and meridional transports of nutrients and plankton and can lead to phytoplankton patchiness at the front. This emphasizes the importance of submesoscale events between interacting mesoscale physical features in the transport of nutrients and plankton, and in explaining the observations. The enhanced phytoplankton concentrations observed during BIOSYNOP are found to be primarily due to advection (convergence) rather than in situ biological growth.     

An Operational Circulation Modeling System for the Gulf of Maine/Georges Bank Region—Part I: Basic Elements

The basic elements of a prototype operational data assimilation modeling system that can provide near-real-time information on the ocean water property and circulation environment in the Gulf of Maine (GOM)/Georges Bank (GB) region are described in this paper. This application of the Harvard Ocean Prediction System (HOPS, Harvard University, Cambridge, MA) model includes development of protocols for the following: 1) the production of model initial fields from an objective blending of climatological and feature model (FM) hydrographic data with fishing-boat-measured bottom temperature data, 2) the ldquowarm startrdquo of the model to produce reasonably realistic initial model fields, 3) converting real-time Fleet Numerical Meteorological and Oceanographic Center (FNMOC, Monterey, CA) model nowcast and forecast winds and/or National Data Buoy Center (NDBC, Stennis Space Center, MS) operational wind measurements to model wind stress forcing fields, and 4) the assimilation of satellite-derived sea surface temperature (SST). These protocols are shown herein to evolve the initial model fields, which were dominated by climatological data, toward more dynamically balanced, realistic fields. Thus, the model nowcasts, with the assimilation of one SST field, are well positioned to produce reasonably realistic ocean fields within a few model days (MDs).     

A comparison between 3DVAR and EnOI techniques for satellite altimetry data assimilation

Two conceptually different assimilation schemes, three dimensional variational (3DVAR) assimilation and Ensemble Optimum Interpolation (EnOI) are compared in the context of satellite altimetric data assimilation. Similarities and differences of the two schemes are briefly discussed and their impacts on the model simulation are investigated.With a tropical Pacific ocean model, two assimilation experiments of sea level anomaly (SLA) data from TOPEX/Poseidon are performed for 5 years from 1997 to 2001. Annual mean states of temperature and salinity fields are compared with analysis data and some independent observations. It is found that EnOI generally produces moderate improvements on both temperature and salinity fields, while changes induced by 3DVAR assimilation are strong and vary remarkably in different areas. For instance, 3DVAR tends to excessively modify the temperature field along the thermocline depth and even deteriorate the simulation, but it is more effective than EnOI below the thermocline depth. However, for the salinity field 3DVAR outperforms EnOI nearly for almost the whole layer. As the difference relative to the WOA01 analysis is compared, it is apparently reduced to below 0.3 psu in most areas in the 3DVAR experiment. On the other hand, the pattern of difference in the EnOI experiment resembles that of the simulation and the magnitude is only diminished to some extent. One advantage of EnOI is that it yields more consistent improvements even in areas where there are large model errors. It is more reliable than 3DVAR in such a sense. It is also revealed that the T–S relation plays a very important role in altimetric data assimilation. Further, the distinct performance of the two schemes can be partly accounted for by their inherent assumptions and settings.     

An experimental analysis for the impact of 3D variation assimilation of satellite data on typhoon track simulation

A series of test simulations are performed to evaluate the impact of satellite-derived meteorological data on numerical typhoon track prediction. Geostationary meteorological satellite (GMS-5) and NOAA‘ s TIROS operational vertical sounder (TOVS) observations are used in the experiments. A three-dimensional variational (3D-Var) assimilation scheme is developed to assimilate the satellite data directly into the Penn State-NCAR nonhydrostatic meteorological model (MM5). Three-dimensional objective analysis fields based on the T213 results and conventional observations are employed as the background fields of the initialization. The comparisons of the simulated typhoon tracks are carried out, which correspond respectively to assimilate different kinds of satellite data. It is found that, compared with the experiment without satellite data assimilation, the 3D-Var assimilation schemes lead to significant improvements on typhoon track prediction. Track errors reduce from approximately 25% at 24 h to approximately 30% at 48 h for 3D-Var assimilation experiments.     

现代海洋/大气资料同化方法的统一性及其应用进展

海洋／大气资料同化的理论基础是用数值模式作为动力学强迫对观测信息进行提炼，或者说，从包含观测误差(噪声)的空间分布不均匀的实测资料中依据动力系统自身的演化规律(动力学方程或模式)来确定海洋／大气系统状态的最优估计。本文对主要的现代海洋／大气资料同化方法，包括最优插值(()ptimal Interpolation，简称()Ⅰ)、变分方法(3—Dimensional Variational和4—Dimensional Variational，分别简称3DVAR和4DVAR)和滤波方法(Filtering)的原理、算法设计和实际应用进行系统地回顾，并对这些资料同化方法的优缺点进行分析和讨论。在滤波框架下，所有的现代资料同化方法都被统一了：()Ⅰ和3DVAR是不随时间变化的滤波器，4DVAR和卡曼滤波是线性滤波器，即非线性滤波的退化情形；而集合滤波能构建非线性的滤波器，因为集合在某种程度上体现了系统的非高斯信息。一个非线性滤波器的主要优点是能计算和应用随时间变化的各阶误差统计距，如误差协方差矩阵。将非线性滤波器计算的随时间变化的误差协方差矩阵引入到（）Ⅰ或4DVAR中，也许能实质性地改进这些传统方法。在实际应用中，方法的优劣可能取决于所选用的数值模式和可获得的计算资源，因此需针对不同的问题选取不同的资料同化方法。由于各种资料同化方法具有统一性，因此可建立测试系统来评价这些方法，从而对各种方法获得更深入的理解，改进现有的资料同化技术，并提高人们对海洋／大气环境的预测能力。     

Method and results of assimilation of climatic data on temperature,salinity, and sea level into a numerical model of the Black Sea

A method for combined assimilation of climatic hydrologic fields of temperature, salinity, and the climatic dynamic level of the Black Sea into a model of sea dynamics is proposed. The monthly mean fields of the dynamic sea level were obtained from the results of assimilation of satellite altimetry data into the model. The statistical characteristics of errors in the forecasts of the level, salinity, and temperature were assumed to be proportional to the statistical characteristics of the differences between monthly mean climatic fields of temperature, salinity, and sea level calculated by means of assimilating altimetry observations of the sea level and analogous climatic hydrologic fields. The climatic fields of currents are reconstructed and analyzed. The assimilation of the climatic altimetry level allows the reproduction (in current fields) of quasi-stationary synoptic anticyclonic eddies located along the periphery of the Black Sea Rim Current.     

Eddy-resolving simulation of plankton ecosystem dynamics in the California Current System

We study the dynamics of the planktonic ecosystem in the coastal upwelling zone within the California Current System using a three-dimensional (3-D), eddy-resolving circulation model coupled to an ecosystem/biogeochemistry model. The physical model is based on the Regional Oceanic Modeling System (ROMS), configured at a resolution of 15 km for a domain covering the entire US West Coast, with an embedded child grid covering the central California upwelling region at a resolution of 5 km. The model is forced with monthly mean boundary conditions at the open lateral boundaries as well as at the surface. The ecological/biogeochemical model is nitrogen based, includes single classes for phytoplankton and zooplankton, and considers two detrital pools with different sinking speeds. The model also explicitly simulates a variable chlorophyll-to-carbon ratio. Comparisons of model results with either remote sensing observations (AVHRR, SeaWiFS) or in-situ measurements from the CalCOFI program indicate that our model is capable of replicating many of the large-scale, time-averaged features of the coastal upwelling system. An exception is the underestimation of the chlorophyll levels in the northern part of the domain, perhaps because of the lack of short-term variations in the atmospheric forcing. Another shortcoming is that the modeled thermocline is too diffuse, and that the upward slope of the isolines toward the coast is too small. Detailed time-series comparisons with observations from Monterey Bay reveal similar agreements and discrepancies. We attribute the good agreement between the modeled and observed ecological properties in large part to the accuracy of the physical fields. In turn, many of the discrepancies can be traced back to our use of monthly mean forcing. Analysis of the ecosystem structure and dynamics reveal that the magnitude and pattern of phytoplankton biomass in the nearshore region are determined largely by the balance of growth and zooplankton grazing, while in the offshore region, growth is balanced by mortality. The latter appears to be inconsistent with in situ observations and is a result of our consideration of only one zooplankton size class (mesozooplankton), neglecting the importance of microzooplankton grazing in the offshore region. A comparison of the allocation of nitrogen into the different pools of the ecosystem in the 3-D results with those obtained from a box model configuration of the same ecosystem model reveals that only a few components of the ecosystem reach a local steady-state, i.e. where biological sources and sinks balance each other. The balances for the majority of the components are achieved by local biological source and sink terms balancing the net physical divergence, confirming the importance of the 3-D nature of circulation and mixing in a coastal upwelling system.     

Reconstruction of subsurface DIC and alkalinity fields in the North Pacific using assimilation of upper ocean data

The subsurface water beneath the summer mixed layer is important to air–sea carbon flux, while its geochemical properties are not frequently observed. A data assimilation method is applied to determine the geochemical fields in the subsurface (i.e., 100 m) from the data collected at the surface in the North Pacific. This method, in the family of the inverse methods, is constructed on a one-dimensional bulk mixed layer model. In addition to temperature and salinity, dissolved inorganic carbon (DIC) and alkalinity are also considered as model variables, whereas biological productivity is omitted. The geochemical properties increase from the fall to the winter, which is the period simulated by the model, as the mixed layer develops and entrains subsurface water rich in DIC and alkalinity. Consequently, the geochemical fields in the mixed layer must have extremely sharp north–south gradients in the western region of the North Pacific and can be reproduced only by enhancing the north–south gradients in the subsurface. The fields reconstructed by the data assimilation provide useful information about the biogeochemical cycles. It is suggested that the large difference in DIC between the surface and the subsurface in the northwestern region is produced by transporting DIC from the mixed layer to the subsurface in the summer, implying extremely high biological productivity. Furthermore, it is suggested that high DIC in the ambient water is maintained by the upwelling of lower layer water.     

Study of long-term variations in the Black Sea fields using an interdisciplinary physical and biogeochemical model

Using an interdisciplinary three-dimensional physical and biogeochemical model developed for the Black Sea, the long-term evolution of marine dynamics and ecosystem is investigated. The hydrophysical fields were calculated from a model of Black Sea circulation with assimilation of hydrographic survey and satellite measurement data from 1971 to 2001. The circulation model reproduces well processes of various scales in both space and time (particularly the seasonal course and interannual variability of main hydrophysical fields). The resulting flow fields are then used to calculate the long-term evolution of the components of the lower level of the food chain in the Black Sea ecosystem. The biogeochemical model used in the calculations is based on the nitrogen cycle and includes a parameterization of the main biological and chemical interactions and processes in the upper layer of the Black Sea. The numerical experiments indicated that the biogeochemical component of the model rather successfully reproduces the main features and evolution trends in the Black Sea ecosystem for the period under consideration: the growth in the phytoplankton biomass during eutrophication and changes in seasonal cycles of the main ecosystem components. Also, the hydrophysical processes were shown to be important for a reliable reproduction of long-term changes in the ecosystem.     

Calculation of adapted black sea fields on the basis of assimilation of climatic temperature and salinity data into the model

A scheme of a periodic assimilation of climatic temperature and salinity data into a model is proposed. In accordance with the criteria chosen on the basis of numerical experiments, an optimum assimilation period of three hours is set. The proposed method allows a hydrodynamic adaptation of the fields of temperature, salinity, and flow velocity that are close to the climatic ones within the framework of the proposed criteria. The main feature of the adapted fields of vertical flows in the layer 0–700 m is their smooth character and an almost complete absence of small-scale noise. The extreme values of vertical velocity in this layer decrease by almost an order of magnitude as compared to analogous values calculated for the variant of data assimilation with a weak adjustment of fields.     

卫星高度计资料在三维海温和盐度数值预报中的应用

随着卫星遥感观测技术的发展,越来越多的卫星观测资料被应用于数值模式的同化研究中.基于国家海洋环境预报中心西北太平洋三维湿盐流预报系统,利用法国CLS中心的沿轨高度计资料的海表面高度异常的融合数据,结合基于三维变分的OVALS(ocean variational analysis system)同化系统,在垂向将海面高度...     

直接同化卫星辐射数据的黄海海雾数值试验研究

基于中尺度大气模式WRF及其3DVAR.模块,采用循环3DVAR数据同化方案,针对6次明显的黄海海雾过程,实施了一系列直接同化ATOVS卫星辐射数据数值试验.在试验中设计了不同化任何观测数据、仅同化GTS常规数据、仅直接同化辐射数据,同时同化二者,以及同化不同疏密程度辐射数据的对比研究方案.利用地面水平能见度与卫星云图对模拟的海雾雾区进行了评估,并比较了各种同化方案所形成初始场的差异.对试验结果的统计分析表明:同化试验较好地再现了影响海雾的天气系统,模拟雾区与实际观测较为吻合,并且初始温度场和湿度场对比不同化任何观测数据的试验有明显的改善;仅同化辐射数据的结果略优于仅同化常规数据的结果,疏化或者只同化海上辐射数据几乎不影响模拟的雾区,但却可以大幅节约计算资源;同时同化常规数据与辐射数据的结果为单独同化它们所得结果的综合体现,总体效果最好.     

Data assimilation for a coastal area morphodynamic model: Morecambe Bay

This paper investigates the use of data assimilation in coastal area morphodynamic modelling using Morecambe Bay as a study site. A simple model of the bay has been enhanced with a data assimilation scheme to better predict large-scale changes in bathymetry observed in the bay over a 3-year period. The 2DH decoupled morphodynamic model developed for the work is described, as is the optimal interpolation scheme used to assimilate waterline observations into the model run. Each waterline was acquired from a SAR satellite image and is essentially a contour of the bathymetry at some level within the inter-tidal zone of the bay. For model parameters calibrated against validation observations, model performance is good, even without data assimilation. However the use of data assimilation successfully compensates for a particular failing of the model, and helps to keep the model bathymetry on track. It also improves the ability of the model to predict future bathymetry. Although the benefits of data assimilation are demonstrated using waterline observations, any observations of morphology could potentially be used. These results suggest that data assimilation should be considered for use in future coastal area morphodynamic models.     

Numerical analysis of hydrophysical fields on the northwestern shelf of the Black Sea

A numerical experiment with assimilation of hydrological observational data from a survey in October 2007 on the northwestern shelf of the Black Sea was carried out using the hydrodynamic model with nonlinear equations of motion, equations of heat and salt advection, and data assimilation. The results of this calculation are compared with thermohydrodynamic fields obtained without taking into account temperature and salinity measurements. It is shown that allowance for the observation data leads to qualitative and quantitative differences in the structure of the hydrophysical fields. Mesoscale eddies and intense jet streams that agree with satellite observations were found in the field of currents and were investigated. These eddies are not resolved in low-resolution field experiments.     

卫星高度计波高资料的同化试验分析

结合经验的台风风场模型和NCEP再分析资料构造了南海台风风场,并以此作为一个第三代海浪模式的输入模拟了1999年约克(York)台风经过南海时所产生的台风浪场.模拟结果显示,尽管文中构造的南海台风风场比较接近真实的风场,但模拟出的台风浪场相对于TOPEX/Poseidon高度计的观测波高仍有一定差异,因此进而采用了一种简单的最优插值同化方法,开展了波高资料的初步同化试验.试验结果显示,资料同化显著地改进了海浪模式的(后)预报精度,而这种改进在涌浪区更加显著,同化有效维持时间大约是43h.     

Data assimilation of partitioned HF radar wave data into Wavewatch III

In this study the assimilation of HF radar data into a high resolution, coastal Wavewatch III model is investigated. An optimal interpolation scheme is used to assimilate the data and the design of a background error covariance matrix which reflects the local conditions and difficulties associated with a coastal domain is discussed. Two assimilation schemes are trialled; a scheme which assimilates mean parameters from the HF radar data and a scheme which assimilates partitioned spectral HF radar data. This study demonstrates the feasibility of assimilating partitioned wave data into a coastal domain. The results show that the assimilation schemes provide satisfactory improvements to significant wave heights but more mixed results for mean periods. The best improvements are seen during a stormy period with turning winds. During this period the model is deficient at capturing the change in wave directions and the peak in the waveheights, while the high sea state ensures good quality HF radar data for assimilation. The study also suggests that there are both physical and practical advantages to assimilating partitioned wave data compared to assimilating mean parameters for the whole spectrum.