大渡河猴子岩水库入库洪水过程预报-实时校正-概率预报集成

与传统确定性预报相比,洪水概率预报能够为防洪调度决策提供更为丰富的信息。以大渡河猴子岩水库以上流域为研究区,建立新安江次洪模型,并采用动态系统响应曲线进行实时洪水预报校正。在确定性预报校正基础上,建立基于水文不确定性处理器（HUP）的次洪概率预报模型,定量分析预报不确定性,实现入库洪水概率预报。结果表明：(1)利用猴子岩流域2009 2019年水文气象资料,建立的新安江次洪模型整体精度较高,率定期和验证期的洪量和洪峰相对误差均在±20%以内,平均确定性系数分别为0.69和0.72;经动态系统响应曲线校正后,洪峰和洪量误差均有降低,率定期和验证期的确定性系数分别提高0.13和0.09。(2)以2020年3场洪水未来48 h预报降雨为输入,新安江模型预报精度不高,且随着预见期增长而降低,但经动态系统响应曲线校正后,整体预报精度有所提高,洪量相对误差减小幅度超50%,确定性系数提高幅度超60%。(3)HUP次洪概率预报模型提供的分布函数中位数Q50的预报精度在一定程度上优于校正后的确定性预报;提供的90%置信区间覆盖率均在90%左右,离散度均在0.40以下,能以相对较窄的区间覆盖大部分实测值...     

基于误差分布估计的三峡水库入库洪水概率预报方法

准确、及时的入库洪水预报,对三峡水库综合效益的发挥和长江流域水旱灾害防御、水资源利用、流域综合管理等具有重要作用。基于预报误差的最优分布估计和分布函数动态参数假定,提出了一种三峡水库入库洪水概率预报方法,并进行了洪水概率预报业务试验。结果表明:本文所提方法科学可行,计算快捷,使用方便,便于在实时作业预报中应用推广;概率预报结果较确定性预报结果,在水量预报、预警效果等方面均有所改善,1～5 d预见期预报的确定性系数提高0.1%～3.4%,水量误差减少0.1%～4.8%,可为三峡水库实时调度提供更可靠的预报信息;所提出的三峡水库入库洪水概率预报业务化产品,可提供更多风险信息,为三峡水库的科学调度,尤其是洪水资源化利用提供更好的优化决策支撑。     

复杂河道洪水预报系统研究——以淮河王家坝至小柳巷区间流域为例

随着社会经济的快速发展,洪水灾害造成的损失日益严重.洪水预报作为一项重要的防洪非工程措施,对防洪、抗洪工作起着至关重要的作用.淮河洪水危害的严重性和洪水演进过程的复杂性使得淮河洪水预报系统的研究长期以来受到高度重视.本文以王家坝至小柳巷区间流域为例,以河道洪水演算为主线,采用新安江三水源模型进行子流域降雨径流预报,概化具有行蓄洪区的干流河道,进行支流与干流、行蓄洪区与干流的洪水汇流耦合计算,采用实时更新的基于多元回归的方法确定水位流量关系,并以上游站点降雨径流预报模型提供的流量作为上边界条件、以下游站点的水位流量关系作为下边界条件,结合行蓄洪调度模型,建立具有行蓄洪区的河道洪水预报系统,再与基于K-最近邻(KNN)的非参数实时校正模型耦合,建立淮河中游河道洪水预报系统.采用多年资料模拟取得了较好的预报效果,并以2003和2007年大洪水为例进行检验,模拟结果精度较高,也证明了所建预报系统的合理性和适用性.     

基于洪水预报系统误差反演的多河段联合校正方法

提出一种基于洪水预报误差系统反演的多河段联合校正方法.采用马斯京根法矩阵方程描述多河段多区间入流的河道汇流过程,基于动力系统反演理论建立洪水预报误差的递推方程,最后利用修正后的多河段状态变量经演算得到预报断面的洪水过程,进而达到多河段联合校正目的.对大渡河上游的应用示例结果表明：多河段联合校正方法考虑了河系中断面间的水力联系及预报误差在时程上的传递规律,可充分利用上游多断面实测和校正信息进行下游预报断面的误差修正,因此具有更高的校正精度和稳定性.     

强烈下渗条件下天然河道洪水演进模拟方法

河道洪水演进模拟是河道洪水预报与汇流计算的主要内容和关键.对于常年断流、河水与地下水长期处于脱节状态的河道,一旦行洪下渗非常强烈,渗漏量很大.不考虑河道下渗洪水演进模拟模型,无法准确模拟真实的洪水运动行为,不能用于洪水模拟预报.本文采用土壤下渗理论描述河道下渗,推导出了河道下渗流量计算公式,建立了基于霍顿下渗公式的河道下渗模拟方法;并把下渗当做单位区间出流,与基于马斯京根康吉法天然河道洪水演进模型进行耦合,构建了强烈下渗条件下天然河道洪水演进模拟模型.针对天然河道水力特性复杂特点,研究了洪水演进模型参数确定方法和波速计算方法.海河流域漳卫河水系岳城水库-蔡小庄段的典型场次洪水应用结果表明,模型能很好反映强烈下渗条件下洪水演进实际情况,具有很高模拟精度,基于土壤下渗理论的下渗量计算公式和模拟方法可以很好地模拟河道下渗过程.这一模型具有计算简便、参数确定简单、适合于洪水预报等特点,在干旱和半干旱地区河道洪水演进模拟预报与河道汇流计算中也具有一定的推广应用价值.     

一次地震预报有效性的"概率统计"评价法--中短期前兆及预报效能评价之二

该提出了评价一次地震预报有效性的概率统计法(套圈模型)。设预报4维空间中一个有限封闭的范围,实际地震落人其范围内即为“报准”,否则为“未报准”。在报准的情况下,预报区间的大小就反映预报水平的高低。这可用预报区域内地震发生的自然概率P来衡量。与“打靶模型”相似,“套圈模型”评价上限是准理想预报尺度,下限是可容忍误差尺度。根据不同的尺度,计算相应的概率。以相应概率对数值之差为比例计算评价值。实际算例表明,此方法与距准误差评价结果相当吻合。     

BMA集合预报在淮河流域应用及参数规律初探

以淮河流域吴家渡水文站作为试验站点,采用基于贝叶斯平均法(BMA)的集合预报模型处理来源于马斯京根法、一维水动力学方法、BPNN(Back Propagation Neural Network)的预报流量序列,通过分析BMA的参数以及其预报结果,对各方法在淮河典型站点流量预报中的适用性进行验证与分析.经2003—2016年19场洪水模拟检验可知,BMA模型能够有效避免模型选择带来的洪水预报误差放大效应,可以提供高精度、鲁棒性强的洪水预报结果.通过进一步比较各模型统计最优的频率与BMA权重值之间的相关性,发现权重值不适用于对单场洪水预报精度评定,而适用于描述多场洪水预报中,模型为最优的统计频率;基于大量先验信息,提前获取BMA的权重等参数,将是指导模型选择、降低洪水预报不确定性、改进洪水预报技术的有效手段.     

基于鲸鱼优化算法的长短期记忆模型水库洪水预报

洪涝灾害是世界主要自然灾害之一,优化洪水预报方案对防洪决策至关重要,然而传统水文模型存在参数多、调参受人为因素影响,泛化能力弱等问题。针对上述问题,本文提出基于改进的鲸鱼优化算法和长短期记忆网络构建自动优化参数的WOA-LSTM模型,通过优化神经网络结构进一步增强该模型的稳定性和精确度,并且建立不同预见期下的洪水预报模型来分析讨论神经网络结构与预报期之间的关系。以横锦水库流域1986—1997年洪水资料为例,其中以流域7个雨量站点的降雨以及横锦站水文资料为输入,不同预见期下洪水过程作为输出,以1986—1993年作为模型的率定期,1994—1997年作为模型的检验期,研究结果表明:(1)以峰现时差、确定性系数、径流深误差和洪峰流量误差作为评价指标,相比较于LSTM模型和新安江模型对检验期的模拟结果表明WOA-LSTM模型拥有更高的精度、预报结果更稳定;(2)结合置换特征值和SHAP法分析模型特征值重要性,增强了神经网络模型的可解释性;(3)通过改变神经网络结构在一定程度避免由于预见期增加和数据关联性下降而导致的模型预报精度下降的问题,最终实验表明该模型在预见期1~6 h下都可以满足横锦水库的洪水预报要求,可以为当地的防洪决策提供依据。     

降雨输入对实时洪水预报精度与预见期的影响

针对降雨输入不确定性对实时洪水预报影响的问题,本文采用不考虑未来预报降雨、考虑未来预报降雨、考虑预报降雨的降雨量误差和降雨时间误差4种方法,以陕西省两个半湿润流域（陈河流域和大河坝流域）为研究区域,分析不同预见期和不同降雨输入情况下洪水预报的精度.研究表明：相对于不考虑未来降雨情况,考虑未来降雨后在预报预见期较长时对预报结果精度提升较大,在预见期较短时对预报结果精度提升不显著;暴雨中心位置不同对预报精度影响也不同,当暴雨中心位于流域下游时降雨量误差对流量预报误差影响更大;降雨量误差主要影响洪量相对误差和洪峰相对误差,且这种影响是线性的,对确定性系数的影响是非线性的二次函数,降雨时间误差主要影响峰现时间误差.     

评价地震预报效能的K指数

本文提出了一种新的评价地震预报效能的方法:K指数方法。它从地震的自然概率、地震预报概率、预报准确系数三个方面评价地震预报效能。文中给出了相应的定量计算方法与实用实例。     

Assessing the reliability of probabilistic flood inundation model predictions

An ability to quantify the reliability of probabilistic flood inundation predictions is a requirement not only for guiding model development but also for their successful application. Probabilistic flood inundation predictions are usually produced by choosing a method of weighting the model parameter space, but previous study suggests that this choice leads to clear differences in inundation probabilities. This study aims to address the evaluation of the reliability of these probabilistic predictions. However, a lack of an adequate number of observations of flood inundation for a catchment limits the application of conventional methods of evaluating predictive reliability. Consequently, attempts have been made to assess the reliability of probabilistic predictions using multiple observations from a single flood event. Here, a LISFLOOD‐FP hydraulic model of an extreme (>1 in 1000 years) flood event in Cockermouth, UK, is constructed and calibrated using multiple performance measures from both peak flood wrack mark data and aerial photography captured post‐peak. These measures are used in weighting the parameter space to produce multiple probabilistic predictions for the event. Two methods of assessing the reliability of these probabilistic predictions using limited observations are utilized; an existing method assessing the binary pattern of flooding, and a method developed in this paper to assess predictions of water surface elevation. This study finds that the water surface elevation method has both a better diagnostic and discriminatory ability, but this result is likely to be sensitive to the unknown uncertainties in the upstream boundary condition. Copyright © 2015 John Wiley & Sons, Ltd.     

Validation of the Flood‐PROOFS probabilistic forecasting system

Probabilistic hydrometeorological forecasting systems are becoming more and more an operational tool used by civil protection centres for issuing flood alerts. One of the most important requests of decision makers is related to the reliability of such systems and to the validation of their predictive performances. For these reasons, this work is devoted to the validation of a probabilistic flood forecasting system called Flood‐PRObabilistic Operational Forecasting System (Flood‐PROOFS). The system is operational in real time, since 2008, in Valle d'Aosta, an alpine Region of northern Italy. It is used by the Civil Protection regional service to issue warnings and by the local water company to protect its facilities. The system manages and uses both real‐time meteorological and satellite data and real‐time data on the operation of the control structures in dam and river, managed by the water company. It has proven a useful tool for flood forecasting and for managing complex situations, facilitating the dialogue between civil protection and the water company during crisis periods. The system uses both a limited area model forecast and a forecast issued by regional expert meteorologists. The main outputs are deterministic and probabilistic discharge forecasts in different outlet areas of the river network. The performance of the system has been evaluated on a 25 months period with different statistical methods such as Brier score and Rank histograms. The results highlight good performances of the system as support system for emitting warnings, but there is a lack of statistics especially for huge discharge events. Copyright © 2013 John Wiley & Sons, Ltd.     

Multi-criteria decision making under uncertainty for flood mitigation

Designs of flood mitigation infrastructural systems are decision-making which are often made under various uncertainties involving multiple criteria. Under the condition of uncertainties, any chosen design alternative has the likelihood to perform inferior to other unselected designs in terms of the adopted performance indicators. This paper introduces a quantitative risk measure based on the concept of expected opportunity loss (EOL) for evaluating the consequence of making the wrong decision. The EOL can be used to assess the relative performance of multiple decision alternatives and is extended to deal with decision problems involving multiple criteria. In particular, the probabilistic features of the consequences associated with a design alternative is considered and used in the Preference Ranking Organization Method of Enrichment Evaluation (PROMETHEE) MCDM technique. The integration of PROMETHEE and decision making under uncertainty is demonstrated through an example of flood damage mitigation planning.     

基于无迹卡尔曼滤波的新安江模型实时校正方法

通过利用实时水文观测数据对洪水预报模型进行校正,可增加流域洪水预报的实时性和精确度.本文讨论了水文模型状态变量选取对滤波效果的影响,并给出了状态变量选取原则.在集总式新安江模型的基础上,结合状态变量选取原则,应用无迹卡尔曼滤波技术构建了新安江模型的实时校正方法.方法应用于闽江邵武流域洪水预报的计算结果表明,采用无迹卡尔曼滤波方法后,不仅能够直接校正模型状态,同时也能有效地提高模型预报精度,适合应用于实际流域洪水预报作业中.     

When does spatial resolution become spurious in probabilistic flood inundation predictions?

Advances in remote sensing have enabled hydraulic models to run at fine scale resolutions, producing precise flood inundation predictions. However, running models at finer resolutions increase their computational expense, reducing the feasibility of running the multiple model realizations required to undertake uncertainty analysis. Furthermore, it is possible that precision gained by running fine scale models is smoothed out when treating models probabilistically. The aim of this paper is to determine the level of spatial complexity that is required when making probabilistic flood inundation predictions. The Imera basin, Sicily is used as a case study to assess how changing the spatial resolution of the hydraulic model LISFLOOD‐FP impacts on the skill of conditional probabilistic flood inundation maps given model parameter and boundary condition uncertainties. We find that model performance deteriorates at resolutions coarser than 50 m. This is predominantly caused by changes in flow pathways at coarser resolutions which lead to non‐stationarity in the optimum model parameters at different spatial resolutions. However, although it is still possible to produce probabilistic flood maps that contain a coherent outline of the flood extent at coarser resolutions, the reliability of these maps deteriorates at resolutions coarser than 100 m. Additionally, although the rejection of non‐behavioural models reduces the uncertainty in probabilistic flood maps the reliability of these maps is also reduced. Models with resolutions finer than 50 m offer little gain in performance yet are more than an order of magnitude computationally expensive which can become infeasible when undertaking probabilistic analysis. Furthermore, we show that using deterministic, high‐resolution flood maps can lead to a spurious precision that would be misleading and not representative of the overall uncertainties that are inherent in making inundation predictions. Copyright © 2015 The Authors Hydrological Processes Published by John Wiley & Sons Ltd.     

Effects of levee cover strength on flood mapping in the case of levee breach due to overtopping

The reliability of a levee system is a crucial factor in flood risk management. In this study we present a probabilistic methodology to assess the effects of levee cover strength on levee failure probability, triggering time, flood propagation and consequent impacts on population and assets. A method for determining fragility curves is used in combination with the results of a one-dimensional hydrodynamic model to estimate the conditional probability of levee failure in each river section. Then, a levee breach model is applied to calculate the possible flood hydrographs, and for each breach scenario a two-dimensional hydrodynamic model is used to estimate flood hazard (flood extent and timing, maximum water depths) and flood impacts (economic damage and affected population) in the areas at risk along the river reach. We show an application for levee overtopping and different flood scenarios for a 98 km reach of the lower Po River in Italy. The results show how different design solutions for the levee cover can influence the probability of levee failure and the consequent flood scenarios. In particular, good grass cover strength can significantly delay levee failure and reduce maximum flood depths in the flood-prone areas, thus helping the implementation of flood risk management actions.

EDITOR D. Koutsoyiannis

ASSOCIATE EDITOR A. Viglione     

On the criteria of model performance evaluation for real-time flood forecasting

Model performance evaluation for real-time flood forecasting has been conducted using various criteria. Although the coefficient of efficiency (CE) is most widely used, we demonstrate that a model achieving good model efficiency may actually be inferior to the naïve (or persistence) forecasting, if the flow series has a high lag-1 autocorrelation coefficient. We derived sample-dependent and AR model-dependent asymptotic relationships between the coefficient of efficiency and the coefficient of persistence (CP) which form the basis of a proposed CE–CP coupled model performance evaluation criterion. Considering the flow persistence and the model simplicity, the AR(2) model is suggested to be the benchmark model for performance evaluation of real-time flood forecasting models. We emphasize that performance evaluation of flood forecasting models using the proposed CE–CP coupled criterion should be carried out with respect to individual flood events. A single CE or CP value derived from a multi-event artifactual series by no means provides a multi-event overall evaluation and may actually disguise the real capability of the proposed model.     

Developing a Web-based flood forecasting system for reservoirs with J2EE / Développement sur Internet avec J2EE d’un système de prévision de crue pour barrages

Abstract

Abstract A flood forecasting system is a crucial component in flood mitigation. For certain important large-scale reservoirs, cooperation and communication among federal, state, and local stakeholders are required when heavy flood events are encountered. The Web-based environment is emerging as a very important development and delivery platform for real-time flood forecasting systems. In this paper, the findings of a case study are presented of the development of a Web-based flood forecasting system for reservoirs using Java 2 platform Enterprise Edition (J2EE). J2EE of Sun Microsystems is chosen as the development solution for the Web-based flood forecasting system, Weblogic 6.0 of BEA as the container provider, and JBuilder 7.0 of Borland as the development tool. One of the key objectives in this project is to establish a collaborative platform for flood forecasting via Web technology in order to render hydrological models and data available to stakeholders and experts involved and thus offer an efficient medium for transferring and sharing information, knowledge and experiences among them. Compared with general Web-based query systems and traditional flood forecasting systems, the Web-based flood forecasting system is more focused on the on-line analysis of model-based forecasting of floods and provides opportunities for improving the transfer of information and knowledge from the hydrological scientists and managers to decision makers. Finally, a prototype system is used to demonstrate the system application.