Using uncertainty and sensitivity analysis for finding the best rainfall-runoff model in mountainous watersheds(Case study: the Navrood watershed in Iran)

In watersheds that have not sufficient meteorological and hydrometric data for simulating rainfall-runoff events, using geomorphologic and geomorphoclimatic characteristics of watershed is a conventional method for the simulation. A number of rainfall-runoff models utilize these characteristics such as Nash-IUH, Clark-IUH, Geomorphologic Instantaneous Unit Hydrograph(GIUH), Geomorphoclimatic Instantaneous Unit Hydrograph(GcIUH), GIUH-based Nash(GIUH-Nash) and GcIUH-based Clark(GcIUH-Clark). But all these models are not appropriate for mountainous watersheds. Therefore, the objective of this study is to select the best of them for the simulation. The procedure of this study is: a) selecting appropriate rainfall-runoff events for calibration and validation of six hybrid models, b) distinguishing the best model based on different performance criteria(Percentage Error in Volume(PEV); Percentage Error in Peak(PEP); Percentage Error in Time to Peak(PETP); Root Mean Square Error(RMSE) and Nash-Sutcliffe model efficiency coefficient(ENS)), c) Sensitivity analysis for determination of the most effective parameter at each model, d) Uncertainty determination of different parameters in each model and confirmation of the obtained results by application of the performance criteria. For application of this procedure, the Navrood watershed in the north of Iran as a mountainous watershed has been considered. The results showed that the ClarkIUH and GcIUH-Clark are suitable models for simulation of flood hydrographs, while other models cannot simulate flood hydrographs appropriately. The sensitivity analysis shows that the most sensitive parameters are the infiltration constant rate and time of concentration in the Clark-IUH model. Also, the most sensitive parameters include the infiltration constant rate and storage coefficient in the GcIUHClark model. The Clark-IUH and GcIUH-Clark models are more sensitive to their parameters. The Latin Hypercube Sampling(LHS) on Monte Carlo(MC) simulation method was used for evaluation of uncertainty of data in rainfall-runoff models. In this method 500 sets of data values are produced and then the peak discharge of flood hydrographs for each produced data set is simulated with rainfall-runoff models. The uncertainty of data changes the value of simulated peak discharge of flood hydrograph. The uncertainty analysis shows that the observed peak discharges of different rainfall-runoff events are within the range of values of simulated by the six hybrid rainfall-runoff models and IUH that inputs of these models were the produced data sets. The range of the produced peak discharge of flood hydrographs by the Clark-IUH and GcIUH-Clark models is wider than those of other models.     

Study on trend prediction and variation on the flow into the longyangxia reservoir

I INTRODUCTIONThe second longest river in China is the Huanghe(Yellow) River on the north of Mt Bayankala on theQinghai-Xizang (Tibet) Plateau. It winds its way through Qinghai, Sichuan, Gansu, Ningxia, Inner Mongolia, Shanxi, Shaanxi, Henan, Shandong, and emptiesinto the Bobal Sea at Kenli County of ShandongProvince. It has a total length of 5464km with adrainage area of 752 443km2. The rushing currentsfrom Longyangxia Gorge to Qingtongxia Gorge in itsupper reaches, running g…     

Full 2D Hydrodynamic Modelling of Rainfall-induced Flash Floods

Mountain catchments are prone to flash flooding due to heavy rainfall. Enhanced understanding of the generation and evolution processes of flash floods is essential for effective flood risk management. However, traditional distributed hydrological models based on kinematic and diffusion wave approximations ignore certain physical mechanisms of flash floods and thus bear excessive uncertainty. Here a hydrodynamic model is presented for flash floods based on the full two-dimensional shallow water equations incorporating rainfall and infiltration. Laboratory experiments of overland flows were modelled to illustrate the capability of the model. Then the model was applied to resolve two observed flash floods of distinct magnitudes in the Lengkou catchment in Shanxi Province, China. The present model is shown to be able to reproduce the flood flows fairly well compared to the observed data. The spatial distribution of rainfall is shown to be crucial for the modelling of flash floods. Sensitivity analyses of the model parameters reveal that the stage and discharge hydrographs are more sensitive to the Manning roughness and initial water content in the catchment than to the Green-Ampt head. Most notably, as the flash flood augments due to heavier rainfall, the modelling results agree with observed data better, which clearly characterizes the paramount role of rainfall in dictating the floods. From practical perspectives, the proposed model is more appropriate for modelling large flash floods.     

基于元胞自动机模型的河道汇流过程模拟

对河道汇流过程进行模拟可为洪水灾害预警预报提供参考。利用水力水文学方法能很好地模拟河道汇流过程,但需要输入的参数多,运算过程复杂,对数据精度要求高,而且在无资料区流域无法确定河道上断面流量情况下,该方法具有一定局限性。本文将元胞自动机模型与水文模型相结合,构建了河道汇流过程中的元胞自动机模型和产流汇流规则。通过建立河道坡面拓扑关系,利用SCS-CN（Soil Conservation Service-Curve Number）模型逐个计算河道元胞上的坡面入流,并利用曼宁方程模拟河道汇流过程,最后在ArcEngine平台下进行二次开发,实现了河道汇流可视化。本文以厦门市茂林溪流域为研究区,对1997年5月6日至7日的一场降雨进行了模拟。将本文模拟结果与该流域其他学者的研究进行了对比分析,结果表明在输入数据与水文模型参数相同的情况下,本文不仅模拟出每次降雨间隔产生的较小洪峰,并且整场降雨产生的最大洪峰流量精度与时间精度均提高了5倍,可以更准确地模拟河道汇流过程,适用于河道汇流可视化,该模拟可以为洪水灾害预警预报提供一定参考。     

下辽河平原浅层地下水脆弱性评价

以下辽河平原为研究对象,在DRASTIC模型基础上,结合RS技术建立了DRASTICL(DRASTIC land use type)模型。利用ArcGIS的水文分析工具对DEM影像进行子流域划分与数据提取。通过对参数进行不确定性表征,对三角模糊参数设定不同α截集,在此基础上将随机参数和模糊参数进行蒙特卡罗模拟。将不同α截集下模拟结果代入模糊模式识别模型,根据累积分布规律,选取不同百分位,从而得出不同α截集与不同百分位地下水脆弱性取值。结合ArcGIS数据可视化表达,得出不同α截集下下辽河平原浅层地下水脆弱性分布图,以此辨析下辽河平原浅层地下水不确定性与脆弱性程度。最后运用灵敏度分析辨别各参数对模拟结果的实际贡献程度。结果表明：（1）模糊模式识别模型用非线性的形式充分反映参数连续性变化对模拟结果产生的影响。（2）加入土地利用类型参数,结果更能反映人类活动影响下地下水脆弱程度。（3）不同α水平、不同百分位、与不同灵敏度系数3个层次的分析有效处理了参数不确定性问题,为制定相关政策提供更加准确的参考依据,对今后本地区的地下水环境开发利用和保护具有重要意义。     

Identification of Suitable Hydrologic Response Unit Thresholds for Soil and Water Assessment Tool Streamflow Modelling

Use of a non-zero hydrologic response unit(HRU) threshold is an effective way of reducing unmanageable HRU numbers and simplifying computational cost in the Soil and Water Assessment Tool(SWAT) hydrologic modelling. However, being less representative of watershed heterogeneity and increasing the level of model output uncertainty are inevitable when minor HRU combinations are disproportionately eliminated. This study examined 20 scenarios by running the model with various HRU threshold settings to understand the mechanism of HRU threshold effects on watershed representation as well as streamflow predictions and identify the appropriate HRU thresholds. Findings show that HRU numbers decrease sharply with increasing HRU thresholds. Among different HRU threshold scenarios, the composition of land-use, soil, and slope all contribute to notable variations which are directly related to the model input parameters and consequently affect the streamflow predictions. Results indicate that saturated hydraulic conductivity, average slope of the HRU, and curve number are the three key factors affecting stream discharge when changing the HRU thresholds. It is also found that HRU thresholds have little effect on monthly model performance, while evaluation statistics for daily discharges are more sensitive than monthly results. For daily streamflow predictions, thresholds of 5%/5%/5%(land-use/soil/slope) are the optimum HRU threshold level for the watershed to allow full consideration of model accuracy and efficiency in the present work. Besides, the results provide strategies for selecting appropriate HRU thresholds based on the modelling goal.     

基于不确定性迭代优化的山地植被遥感制图

山地因其较高的异质性和特殊的环境特征给遥感科学及其应用带来了诸多问题和挑战。为实现山地植被信息的精准提取,本研究选择部分滇西北山地区域作为研究区开展方法实验,利用高分辨率遥感影像数据和数字高程模型,结合分区分层感知思想,提出一种基于不确定性理论的山地植被型组分类制图方法。首先结合地形对研究区影像进行多尺度分割制作图斑;然后根据图斑特征使用随机森林方法进行分类,将分类结果与对应类别样本间的相似性作为优化目标,并构建混合熵模型定量计算图斑推测类型的不确定性,据此进行针对性的样本补充和分类模型的迭代优化。实验总体分类精度达90.8%,较迭代前提升了29.4%,Kappa系数达到0.875。在高不确定性区域,该方法相比使用一次性补样和随机补样方法的分类结果,精度分别提高了17%和13%。研究结果表明,通过人机交互的方式,基于不确定性理论为样本库融入增量信息的迭代优化方法能够有效提高植被型组分类的精度,相较于传统的样本选择方法具有更高的效率和更低的不确定性。     

Forecast of Flood in Chaohu Lake Basin of China Based on Grey-Markov Theory

Flood is one kind of unexpected and the most common natural disasters, which is affected by many factors and has complex mechanism. At home and abroad, there is still no mature theory and method used for the long-term forecast of natural precipitation at present. In the present paper the disadvantages of grey GM (1, 1) and Markov chain are ana- lyzed, and Grey-Markov forecast theory about flood is put forward and then the modifying model is developed by making prediction of Chaohu Lake basin. Hydrological law was conducted based on the theoretical forecasts by grey system GM (1, 1) forecast model with improved Markov chain. The above method contained Stat-analysis, embodying scientific approach, precise forecast and its reliable results.     

降雨数据空间分辨率在城市流域洪峰变化分析中的影响

在气候变化和城市化背景下,城市暴雨的局部特征愈发明显,而降雨监测和预报数据空间分辨率难以满足空间分布描述要求,造成城市洪涝模拟和预测结果存在不确定性。本文选择深圳市城区的3个流域,基于118场实测降雨数据,通过逐步增大网格尺寸以达到降低降雨空间分辨率的效果,同时以流域边界作为限制,保证流域内面平均雨量不变,然后将不同空间分辨率的降雨数据分别作为城市洪涝模型的输入条件,对流域出口断面的洪水过程进行模拟计算,每场降雨得到6种不同空间分辨率下的洪峰流量。结合降雨中心、流域几何中心等空间特征参数,分析降雨空间分辨率对流域洪峰的影响。研究结果表明,针对118场实测降雨,随着降雨数据空间分辨率逐渐降低,3个流域的出口断面的洪峰变化幅度逐渐增大;如果忽略了流域内部降雨数据的空间分布差异,洪峰变化幅度和方向与降雨中心和流域几何中心的空间关系具有相关性,降雨中心在流域几何中心上游方向的距离越大,洪峰降低的幅度越大。本文基于3个城市流域得到了相似的研究结论,研究方法可为城市流域洪涝模拟提供借鉴,研究结果可为河道洪水风险评价提供参考。     

GRAPES-Meso模式动力框架与物理过程对预报误差影响研究

为了研究GRAPES-Meso区域中尺度模式误差特点,评估模式动力框架和物理过程对预报误差的影响重要性,为GRAPES区域集合预报系统方案设计提供参考,基于GRAPES中尺度模式设计了4组对比试验,每组试验对2008年3个不同类型天气过程进行了数值模拟,获得如下结论:(1)GRAPES-Meso模式存在较为显著的系统性误差,系统性误差水平和垂直分布特征主要由GRAPES模式动力框架产生,物理过程对系统性误差影响相对较小;(2)在模式层底和模式层顶,GRAPES模式层与等压面层转换方案中,预报存在较为明显的垂直插值误差;(3)边界层方案对GRAPES模式低层动力场预报误差有重要影响,可以显著减少模式低层动力场预报误差。结果表明减少动力框架预报误差是改进GRAPES-Meso模式的重点,在GRAPES-Meso集合预报系统的设计中,需要重点考虑动力框架引起的模式不确定性。     

Assimilation of ASAR data with a hydrologic and semi-empirical backscattering coupled model to estimate soil moisture

The most promising approach for studying soil moisture is the assimilation of observation data and computational modeling.However,there is much uncertainty in the assimilation process,which affects the assimilation results.This research developed a one-dimensional soil moisture assimilation scheme based on the Ensemble Kalman Filter(EnKF)and Genetic Algorithm(GA).A two-dimensional hydrologic model-Distributed Hydrology-Soil-Vegetation Model(DHSVM)was coupled with a semi-empirical backscattering model(Oh).The Advanced Synthetic Apertture Radar(ASAR)data were assimilated with this coupled model and the field observation data were used to validate this scheme in the soil moisture assimilation experiment.In order to improve the assimilation results,a cost function was set up based on the distance between the simulated backscattering coefficient from the coupled model and the observed backscattering coefficient from ASAR.The EnKF and GA were used to re-initialize and re-parameterize the simulation process,respectively.The assimilation results were compared with the free-run simulations from hydrologic model and the field observation data.The results obtained indicate that this assimilation scheme is practical and it can improve the accuracy of soil moisture estimation significantly.     

四川盆地汛期旱涝预测研究

对近１０年来四川盆地汛期降水预测水平进行了定量评估,并与国家气候中心预测水平作了对比。在分析影响四川盆地汛期降水主要物理因素的基础上,提出了四川盆地汛期旱涝预测的基本思路和技术方法,并在实际业务预报中取得了较好的效果。     

基于LBS和深度学习的旅游景区客流量的高时频预测

为实现精准的旅游景区客流量的高时频预测,本研究构建了一套基于LBS和深度学习模型的预测方法。此方法可通过对LBS数据的转换实现预测的空间范围与时频控制,并通过方法的核心模型——基于双向循环神经网络和GRU算法构建的深度双向GRU(DBi-GRU)模型完成预测。为检验方法的有效性,研究以深圳大梅沙海滨公园为例对方法进行实验测试。实验使用拟合曲线、误差指标及DM检验3种方法评估DBi-GRU模型的预测效果。此外,实验还设置了其他五种深度学习模型作为DBi-GRU的对照模型,测试基于不同深度学习算法的模型之间的预测水平差异。实验结果表明:(1)本研究提出的DBi-GRU模型在景区客流量高时频预测中具有理想的预测效果,在高峰时段的客流量预测方面也具有较高准确性,预测效果明显优于其他深度学习模型;(2)基于双向循环网络的模型的效果普遍优于基于常规循环网络的模型。尤其是基于双向LSTM算法的模型,虽然预测的准确度略逊色于DBi-GRU模型,但在模型性能上与其的差异并不显著;(3)在相同网络参数下,GRU算法较前人采用的LSTM和RNN算法有着更高的预测准确性。本研究为客流量预测领域的研究提供了一种...     

Development of an oil spill forecast system for offshore China

An oil spill forecast system for of fshore China was developed based on Visual C++. The oil spill forecast system includes an ocean environmental forecast model and an oil spill model. The ocean environmental forecast model was designed to include timesaving methods, and comprised a parametrical wind wave forecast model and a sea surface current forecast model. The oil spill model was based on the "particle method" and fulfi lls the prediction of oil particle behavior by considering the drifting, evaporation and emulsifi cation processes. A specifi c database was embedded into the oil spill forecast system, which contained fundamental information, such as the properties of oil, reserve of emergency equipment and distribution of marine petroleum platform. The oil spill forecast system was successfully applied as part of an oil spill emergency exercise, and provides an operational service in the Research and Development Center for Off shore Oil Safety and Environmental Technology.     

青岛市大沽河流域防汛信息系统

随着计算机信息技术在水利学科中应用的深入,以及数字水利、数字流域概念的提出,水利信息化已经成为这一行业领域的主要发展方向,各种流域模型在与地理信息系统平台结合后在数据的收集、存储,提取与处理的基础上都有了很大的发展,在水利工程的规划设计、防洪减灾、水量分配等方面有更为广泛的应用。青岛市大沽河流域内洪水有来势猛、峰高量大和暴涨暴落的特性,为满足防汛管理部门对洪水预测及相关信息快速浏览查询的需求,建立了青岛市大沽河流域防汛信息系统。该系统在MAPX地理信息系统开发软件、数据库存储技术和水流数学模型的支持下,实现了大沽河流域的地理信息、水文信息以及其他信息的综合管理,同时通过水流数学模型对一维和二维洪水进行预测预报,并根据用户的需要完成所需的查询与分析功能,以最直观简洁的方式向用户展示系统的输出结果。系统数据库的开发在满足系统要求的基础上完成了与青岛市雨情水情自动监测系统的对接,保证了水文数据的实时性、完整性和准确性。系统集成了一维非恒定流模型与平面二维非恒定流水流模型,可以根据大沽河流域降雨量计算干流的水位变化以及发生溃堤后洪水的淹没过程。应用表明,该系统界面友好、功能完善、性能良好,具有很强的实用性。     

Effect of Uncertainty of Grid DEM on TOPMODEL： Evaluation and Analysis

1 Introduction Digital Elevation Model (DEM) is a digital representa- tion of the earth’s surface and is typically represented in three formats: Grid DEM, TIN (Triangular Irregular Network) and contour-based storage structure. Being readily available and easily operating, Grid DEM has been widely applied to the analysis of hydrological problems (Moore et al., 1991) and other fields. Unfor- tunately, DEM has its inherent errors, the departure of a measurement from its true value, whic…     

APPROACH OF DETERMINISTIC GEOMORPHOLOGIC INSTANTANEOUS UNIT HYDROGRAPH TO HYDROLOGICAL PROCESSES‘ SIMULATION IN KARST AREA

In order to analyze the condition of special landform and regularity of special hydrological movement inkarst area, the method of Geomorphologic Instantaneous Unit Hydrograph (GIUH) has been used to divide the river net-works into three grades in the research area and the karst hydrologic processes were simulated with Nash instantaneous u-nit hydrograph(NIUH). The combination of the GIUH and NIUH fully considers the moving path of water particles indrainage and the regional composition of runoff at the end of the drainage, resulting in a rational combination of probabili-ty of GIUH and NIUH and geomorphologic parameters which could reflect the complexity of the landform structure ofkarst drainage. The results showed that the combined method of GIUH and NIUH has clear physical concept and accept-able precision, which can be widely applied on hydrological studies of karst area.     

Assessment of the performance of WEPP in purple soil area with simulated rainfall experiments

The water erosion prediction project(WEPP) model is a popular water erosion prediction tool developed on the basis of the physical processes of water erosion.Although WEPP has been widely used around the world,its application in China is still insufficient.In this study,the performance of WEPP used to estimate the runoff and soil loss on purple soil(Calcaric Regosols in FAO taxonomy) sloping cropland was assessed with the data from runoff plots under simulated rainfall conditions.Based on measured soil properties,runoff and erosion parameters,namely effective hydraulic conductivity,inter-rill erodibility,rill erodibility,and critical shear stress were determined to be 2.68 mm h-1,5.54 × 106 kg s-1 m 4,0.027 s m 1 and 3.5 Pa,respectively,by using the recommended equations in the WEPP user manual.The simulated results were not good due to the low Nash efficiency of 0.41 for runoff and negative Nash efficiency for soil loss.After the four parameters were calibrated,WEPP performed better for soil loss prediction with a Nash efficiency of 0.76.The different results indicated that the equations recommended by WEPP to calculate parameters such as erodiblity and critical shear stress are not suitable for the purple soil areas,Sichuan Province,China.Although the predicted results can be accepted by optimizing the runoff and erosion parameters,more research related to the determination of erodibility and critical sheer stress must be conducted to improve the application of WEPP in the purple soil areas.     

Sensitivity evaluation of the flash flood warning system introduced to ungauged small mountainous basins in Korea

This study evaluates the uncertainty involved in the determination of the flash flood guidance(FFG) of the flash flood warning system(FFWS) for a small mountainous region(FFWS_MR)in Korea. The sensitivity is evaluated both at each step to determine the FFG and for all steps together. The results show that the relative difference of the FFG is about 50% of the current system, most of which involves the channel width. Especially, the use of some specific empirical equations to estimate the major parameters results in a considerable amount of the relative difference of the FFG. In addition, though only four basins were field-surveyed, it is found that the indirect estimation of the major parameters always introduces some amount of additional uncertainty. In conclusion, accurate estimation of the major parameters must be the most important procedure to derive an accurate FFG, among which the channel cross-section at the exit of the basin lies at the center.     

因子集成及综合相似预测模型的研究和应用

针对西南区域（云、贵、川、渝）汛期（６－８月）旱涝预测问题,建立了一种以多种物理因子为基础的西南区域降水量预测模型,即因子集成及综合相似预测模型。该模型既考虑影响西南区域汛期降水的主要物理因素变化,又注重区域要素场自身演变规律。对于影响汛期旱涝的物理因子,采用对多个因素进行综合集成的办法,来体现所有因素对旱涝影响的整体作用。为了强调预测结果的客观性,我们借鉴目前ＮＷＰ集合预报思想,引入综合相似方法,给出西南区域大范围汛期降水距平百分率预测结果。近年来业务试用表明,该方法预测准确率稳定,有明显的预报技巧