KALMAN FILTERING CORRECTION IN REAL-TIME FORECASTING WITH HYDRODYNAMIC MODEL

Accurate and reliable flood forecast is crucial for efficient real-time river management, including flood control, flood warning, reservoir operation and river regulation. In order to improve the estimate of the initial state of the forecasting system and to reduce the errors in the forecast period a data assimilation procedure was often need. The Kalman filter was proven to be an efficient method to adjust real-time flood series and improve the initial conditions before the forecast. A new model integrating the hydraulic model with the Kalman filter for real-time correction of flood forecast was developed and applied in the Three Gorges interzone of the Yangtze River. The method was calibrated and verified against the observed flood stage and discharge during Three Gorges Dam construction periods （2004）. The results demonstrate that the new model incorporates an accurate and fast updating technique can improve the reliability of the flood forecast.     

An integrated measurement and modeling methodology for estuarine water quality management

This paper describes research undertaken by the authors to develop an integrated measurement and modeling methodology for water quality management of estuaries. The approach developed utilizes modeling and measurement results in a synergistic manner. Modeling results were initially used to inform the field campaign of appropriate sampling locations and times, and field data were used to develop accurate models.Remote sensing techniques were used to capture data for both model development and model validation. Field surveys were undertaken to provide model initial conditions through data assimilation and determine nutrient fluxes into the model domain. From field data, salinity relationships were developed with various water quality parameters, and relationships between chlorophyll a concentrations, transparency, and light attenuation were also developed. These relationships proved to be invaluable in model development, particularly in modeling the growth and decay of chlorophyll a. Cork Harbour, an estuary that regularly experiences summer algal blooms due to anthropogenic sources of nutrients,was used as a case study to develop the methodology. The integration of remote sensing, conventional fieldwork, and modeling is one of the novel aspects of this research and the approach developed has widespread applicability.     

Modeling water resources under competing demands for sustainable development: A case study of Kaligandaki Gorge Hydropower Project in Nepal

The challenges posed by climate change require that the quantity and quality of water resources in Nepal be managed with sustainable development practices. The communities around the Kaligandaki Gorge Hydropower Project in the Myagdi District of Nepal depend on river flow for most of their rural and agricultural needs. Without a sustainable development plan, the growing population of the region, confined in an area with declining water resources, will face serious challenges to economic growth. Meteorological data show increasing annual average rainfall at a slight rate of about 0.284 mm/year, with erratic annual percentage change in rainfall in the area. The mean and minimum temperatures show decreasing trends at the rates of 0.05℃ and 0.140 C per year, respectively. An assessment of the impacts on water availability for domestic and irrigation usage in the face of competing demands caused by the hydropower development project in the Kaligandaki Gorge was undertaken. The water demand and supply modeling were conducted using the water evaluation and planning(WEAP) model, based on discharge data from the Kaligandaki River, which were obtained from the Department of Hydrology and Meteorology, Nepal. The available data from 2001 to 2003 were used to estimate the model parameters while the stability of these parameters was tested with a validation period from2004 to 2007. The performance of the model was assessed through statistical measures of calibration with the root mean square error and coefficient of determination, whose values were 0.046% and 0.79, respectively. Two scenarios were created in addition to the base case scenario:the discharge decrement scenario and new irrigation technology scenario. Analysis showed that a prioritization of demands will be necessary in the area in the near future for the purpose of sustainability of water resources, due to climate change impacts.     

COMPUTATIONAL RESEARCH ON WAVE MAKING OF MOVING WIGLEY HULL IN TIME DOMAIN

Based on Green＇s theorem, a time domain numerical model was constructed to simulate wave making phenomenon caused by a moving ship. In this article, the Rankine sources and dipoles were placed on boundary surfaces （i.e., the ship surface and free surface）, and a time-stepping scheme was employed. Its unique characteristic is that steady state can be realized from initial value by employing the time-stepping scheme and unsteady free surface conditions. In time domain, if the results of unsteady flow problem tend to data stabilization after many time steps of computation, they could be regarded as the data of steady ones. This model could be employed to steady or unsteady problems. Theoretical reasoning and computational process of this method was described in detail The linear and nonlinear boundary conditions on body surface were studied, and the relative means to realize these boundary conditions in iterative computation were also discussed. Some proper parameters about the model of the Wigley hull were determined by many numerical tests, and their influences on wave making resistance and wave pattern were discussed. According to the comparison between numerical results and data available in relative references, the method used in this work is proven to be a reliable method in time domain. And the lattice reorganization in every time step computation is a feasible numerical approach.     

Investigation of fluid flow-induced particle migration in granular filters using a DEM-CFD method

In embankments and earth dams, the granular filter used to protect the base soil from being eroded by the fluid flow is a major safety device. In this paper, the migration mechanism of the base soil through this type of filters with a fluid flow in the base soil-filter system is studied by using the coupled distinct element method and computational fluid dynamics（DEM-CFD） model. The time-dependent variations of the system parameters such as the total eroded base soil mass, the distribution of the eroded particles within the filter, the porosity, the pore water pressure, and the flow discharge are obtained and analyzed. The conceptions of the trapped particle and the trapped ratio are proposed in order to evaluate the trapped condition of the base soil particles in the filter. The variation of the trapped ratio with time is also analyzed. The results show that the time evolutions of the parameters mentioned above are directly related to the gradation of the filter, which is defined as the representative particle size ratio of the base soil to the filter using an empirical filter design criterion. The feasibility of the model is validated by comparing the numerical results with some experimental and numerical results.     

UNIFIED MODEL FOR SPLASH DROPLETS AND SUSPENDED MIST OF ATOMIZED FLOW

In this article, the unified mathematical model for splash droplets and suspended mist of atomized flow was established, which classifies the atomized sources into the splash source and the suspended source. For the splash source, the Lagrangian method was used to simulate the random motion of splash water droplets, and for the suspended source the theory of air-water two-phase flow was used to simulate the mist flow moving in particle clouds. The rainfall intensity of the atomized flow was obtained by summarizing the rainfall intensities relative to the above two types of atomized sources. Both experimental data and prototype observation data were used for the verification of the mathematical model. For both the distribution of rainfall intensity, and the outer edge of the atomized flow, the simulation results are in agreement with the experimental data or prototype observation data.     

ONE-AND TWO-DIMENSIONAL COUPLED HYDRODYNAMICS MODEL FOR DAM BREAK FLOW

1-D and 2-D mathematical models for dam break flow were established and verified with the measured data in laboratory. The 1-D and 2-D models were then coupled, and used to simulate the dam break flow from the reservoir tail to the dam site, the propagation of dam break waves in the downstream channel, and the submergence of dam break flow in the downstream town with the hydrodynamics method. As a numerical example, the presented model was employed to simulate dam break flow of a hydropower station under construction. In simulation, different dam-break durations, upstream flows and water levels in front of dam were considered, and these influencing factors of dam break flow were analyzed, which could be referenced in planning and designing hydropower stations.     

A TVD Scheme for Incompressible Flow Coupled With Different Turbulence Modles on A Ground-Mounted Square-RIB Flow

A finite-difference Total Variation Diminishing (TVD) numerical simulation model for coupling the Reynolds Averaged Navier-Stokes (RANS) equations, pressure-relative continuity equation and various k-ε turbulence models was developed to solve the incompressible flow based on the pseudo-compressibility method. The hyperbolicity of all these equations was studied and the discretization of the fully coupling equations with all the primal variables and source terms were made in this article. Numerical simulation for modeling the flow around a ground-mounted square rib was implemented and validated by comparing with the published wind tunnel experimental data. It is shown that such a numerical simulation method with a proper turbulence model has a very good accuracy to simulate the flow around a surface-mounted rib. It is concluded that the Renormalization Group (RNG) and Chen-Kim k-ε turbulence models have much better ability to predict the characteristics of the vortex structure and flow separation than the standard k-ε model.     

A TVD SCHEME FOR INCOMPRESSIBLE FLOW COUPLED WITH DIFFERENT TURBULENCE MODLES ON A GROUND- MOUNTED SQUARE-RIB FLOW

A finite-difference Total Variation Diminishing (TVD) numerical simulation model for coupling the Reynolds Averaged Navier-Stokes (RANS) equations, pressure-relative continuity equation and various k-εturbulence models was developed to solve the incompressible flow based on the pseudo-compressibility method. The hyperbolicity of all these equations was studied and the discretization of the fully coupling equations with all the primal variables and source terms were made in this article. Numerical simulation for modeling the flow around a ground-mounted square rib was implemented and validated by comparing with the published wind tunnel experimental data. It is shown that such a numerical simulation method with a proper turbulence model has a very good accuracy to simulate the flow around a surface-mounted rib. It is concluded that the Renormalization Group (RNG) and Chen-Kim k-εturbulence models have much better ability to predict the characteristics of the vortex structure and flow separation than the standard k-εmodel.     

基于 EnKF 的新安江模型参数和变量同步估计方法

数据同化方法可提高数值预报的时效性和准确性,且该方法已在水文领域得到应用,并得到快速发展。为了提高新安江模型径流模拟预报精度,采用集合卡尔曼滤波方法同化径流数据,对参数和状态变量进行同步校正估计。通过对三水源新安江模型进行理想条件下的数值实验,在同时考虑模型自身、模型参数以及观测数据的不确定性的情况下,分析了参数均值和方差改变、集合大小、同化参数的敏感性以及相关性分析对同化过程的影响。结果表明:集合卡尔曼滤波算法具有可行性,且参数均值越接近真值、方差适当增加,集合大小适中,同化参数敏感性较低以及参数与变量间相互独立时,能在一定程度上增加径流同化精度。该研究可为同类型参数同化估计提供一定参考依据。     

基于EnKS和SWAT模型的闽江流域径流数据同化

地表水文过程中观测变量对状态变量的响应存在时间滞后性,为提高径流数据同化的精度,以闽江流域为研究区,基于集合卡尔曼平滑器（EnKS）和SWAT模型,构建径流数据同化方案,并与集合卡尔曼滤波（EnKF）方法进行对比,评价不同同化模型的精度,分析数据同化对不同径流分量的影响。结果表明：EnKS最优时间窗口长度在不同水文周期、流域存在差异;考虑水文模型的时间滞后性可以有效提高模型的同化精度,对比EnKF方法,EnKS方法的纳什效率系数(NSE)在七里街、沙县、竹岐3个站点上分别提升了0.03、0.12、0.03,均方根误差(RMSE)分别减小了7.43%、26.81%、4.25%;数据同化方法对不同径流分量的改进程度存在空间异质性和时间异质性,在高渗透率土壤和陡坡区域EnKS方法能使壤中流获得更显著的改进,丰水期EnKS方法对地表径流的改进较枯水期更明显。     

非饱和土壤水的集合卡尔曼滤波Ⅰ:状态向量与非饱和流算法的选择

土壤水运动是水分循环中的基本过程,但土壤水预测面临着参数获取难、预测精度差等挑战。数据同化技术为土壤水参数估计和精确预报提供了一种新的方法。本文建立了基于3种不同非饱和水流求解方法的集合卡尔曼滤波(En KF)算法,针对状态向量的选择和正演模型的选择两个问题,研究了非饱和土壤水En KF的计算性能。研究结果表明:对于非线性非饱和水流问题,同时更新水头和参数比仅仅更新水头能够取得更好的预测效果,特别是当多参数未知时;En KF本质上是Monte Carlo方法,极端样本容易导致Picard-h和Picard-mix算法的崩溃,因此传统的HYDRUS程序与复杂非饱和土壤水的数据同化兼容性不佳;当同时同化水头和参数时,如果极端的样本值能够快速得以更新,Picard-h和Picard-mix算法在数据同化模拟中的适用性能得以提升;但由于观测信息对参数的校正能力取决于特定的问题和条件,Ross算法是执行非饱和土壤水数据同化模拟的更好选择。     

Application of ensemble H-infinity filter in aquifer characterization and comparison to ensemble Kalman filter

Though the ensemble Kalman filter (EnKF) has been successfully applied in many areas, it requires explicit and accurate model and measurement error information, leading to difficulties in practice when only limited information on error mechanisms of observational instruments for subsurface systems is accessible. To handle the uncertain errors, we applied a robust data assimilation algorithm, the ensemble H-infinity filter (EnHF), to estimation of aquifer hydraulic heads and conductivities in a flow model with uncertain/correlated observational errors. The impacts of spatial and temporal correlations in measurements were analyzed, and the performance of EnHF was compared with that of the EnKF. The results show that both EnHF and EnKF are able to estimate hydraulic conductivities properly when observations are free of error; EnHF can provide robust estimates of hydraulic conductivities even when no observational error information is provided. In contrast, the estimates of EnKF seem noticeably undermined because of correlated errors and inaccurate error statistics, and filter divergence was observed. It is concluded that EnHF is an efficient assimilation algorithm when observational errors are unknown or error statistics are inaccurate.     

基于多源遥感土壤湿度与模型数据同化的流域径流模拟

为提升径流模拟精度,以秦淮河流域为例,采用集合平均法将SMAP、SMOS、AMSR2卫星遥感土壤湿度融合并利用地形湿度指数进行空间降尺度处理,采用卡尔曼滤波算法和栅格新安江模型进行遥感融合土壤湿度同化。对2016—2018年秦淮河流域3个流量站记录的11场洪水进行模型数据同化的结果表明：日尺度率定期洪峰、径流深相对误差合格率均为71.43%,验证期洪峰、径流深相对误差合格率分别为66.67%和100%;经同化后,8场洪水径流深误差减小,平均误差降低29.01%;8场洪水确定性系数增大,范围在0.01～0.09之间,模拟精度最高可提升11.84%;同化多源遥感土壤湿度能有效改善土壤湿度估计的准确性,进而提升径流模拟精度。     

基于GPU和数据同化的深水湖库水温与溶解氧中短期预报

深水水库通常存在季节性温度分层,由温度分层引起溶解氧等水质指标的分层还会诱发库区水环境水生态问题。当前在中短期时间尺度上对水库水温和溶解氧进行预报的研究相对较少,提高数学模型的模拟效率与精度对提升中短期预报效果至关重要。本文采用集合卡尔曼滤波算法作为同化方法,基于CE-QUAL-W2模型建立水库水动力水质数学模型,基于OpenACC的GPU并行方法提升模型计算效率,构建大黑汀水库水温与溶解氧的数据同化系统,在中短期时间尺度上开展水库水温与溶解氧高精度、高效率预报。预报结果符合水库水温与溶解氧的中短期变化规律,能够为大黑汀水库的供水与生态安全提供技术支撑。     

Analysis of influence of observation operator on sequential data assimilation through soil temperature simulation with common land model

An observation operator is a bridge linking the system state vector and observations in a data assimilation system. Despite its importance, the degree to which an observation operator influences the performance of data assimilation methods is still poorly understood. This study aimed to analyze the influences of linear and nonlinear observation operators on the sequential data assimilation through soil temperature simulation using the unscented particle filter (UPF) and the common land model. The linear observation operator between unprocessed simulations and observations was first established. To improve the correlation between simulations and observations, both were processed based on a series of equations. This processing essentially resulted in a nonlinear observation operator. The linear and nonlinear observation operators were then used along with the UPF in three assimilation experiments: an hourly in situ soil surface temperature assimilation, a daily in situ soil surface temperature assimilation, and a moderate resolution imaging spectroradiometer (MODIS) land surface temperature (LST) assimilation. The results show that the filter improved the soil temperature simulation significantly with the linear and nonlinear observation operators. The nonlinear observation operator improved the UPF's performance more significantly for the hourly and daily in situ observation assimilations than the linear observation operator did, while the situation was opposite for the MODIS LST assimilation. Because of the high assimilation frequency and data quality, the simulation accuracy was significantly improved in all soil layers for hourly in situ soil surface temperature assimilation, while the significant improvements of the simulation accuracy were limited to the lower soil layers for the assimilation experiments with low assimilation frequency or low data quality.     

湖泊藻类动态模型数据同化模式的改进

以太湖为研究区域,采用2014—2016年的水环境生态监测数据,率定了三维水生态动力学模型(3DHED),模拟了太湖蓝藻生物量的时空变化;通过融入遥感数据建立了基于集合卡尔曼滤波(EnKF)的蓝藻生物量预测数据同化(DA)模式,同时提出了一种改进数据同化(mDA)的策略,降低了遥感数据不确定性的影响,显著提升了模型模拟精度。结果表明:相比3DHED蓝藻生物量的模拟结果,DA模拟结果的均方根误差均值降低了10.4%,IOA均值增加了48.8%;mDA在DA基础上对蓝藻生物量的模拟精度进一步提升,其均方根误差均值为1.16 mg/L,在DA基础上降低了8.6%,IOA均值为0.71,在DA基础上增加了10.9%,并有效提升了对蓝藻生物量峰值的捕捉能力,表明提出的mDA方法能有效减小原DA模式中遥感观测数据误差的影响,提升水华模拟精度。     

Formulation of the Evolutionary-Based Data Assimilation, and its Implementation in Hydrological Forecasting

Multi-objective evolutionary algorithms (MOEAs) have gained popularity for their capability to handle complex and nonlinear problems. MOEAs are population-based search tools which employ the concept of biological evolution and natural selection. While MOEAs have been applied in numerous hydrological studies for parameter estimation, their formulation for solving data assimilation (DA) problems has not been completely formalized in the literature. This study presents the evolutionary-based data assimilation (EDA) where it formulates the MOEA strategy into an applied DA procedure. The study outlines the stochastic and adaptive capabilities of MOEAs, and shows how MOEA operators including Pareto dominance, crossover, and random variation are naturally suited to handle DA problems. The EDA employs the cost function from variational DA to approximate the least squares estimate between ensemble simulations and perturbed observation. The EDA uses the MOEA strategy to evolve a population of competing members through several cycles of evolution at each assimilation step. The EDA determines several non-dominated members for each assimilation time step, allows these members to evolve, and evaluate updated members for subsequent time steps. Several ensemble members are evaluated for each assimilation time step but the updated ensembles are determined as a subset of the final evolved population which comprise the Pareto-optimal set. The EDA has been illustrated in a practical implementation to assimilate daily streamflow into the Sacramento Soil Moisture Accounting model.