Modeling Flow and Sediment Transport in a River System Using an Artificial Neural Network

A river system is a network of intertwining channels and tributaries, where interacting flow and sediment transport processes are complex and floods may frequently occur. In water resources management of a complex system of rivers, it is important that instream discharges and sediments being carried by streamflow are correctly predicted. In this study, a model for predicting flow and sediment transport in a river system is developed by incorporating flow and sediment mass conservation equations into an artificial neural network (ANN), using actual river network to design the ANN architecture, and expanding hydrological applications of the ANN modeling technique to sediment yield predictions. The ANN river system model is applied to modeling daily discharges and annual sediment discharges in the Jingjiang reach of the Yangtze River and Dongting Lake, China. By the comparison of calculated and observed data, it is demonstrated that the ANN technique is a powerful tool for real-time prediction of flow and sediment transport in a complex network of rivers. A significant advantage of applying the ANN technique to model flow and sediment phenomena is the minimum data requirements for topographical and morphometric information without significant loss of model accuracy. The methodology and results presented show that it is possible to integrate fundamental physical principles into a data-driven modeling technique and to use a natural system for ANN construction. This approach may increase model performance and interpretability while at the same time making the model more understandable to the engineering community.     

A COMPARATIVE STUDY OF LINEAR AND NONLINEAR TIME SERIES MODELS FOR WATER QUALITY1

ABSTRACT: Surface water quality data are routinely collected in river basins by state or federal agencies. The observed quality of river water generally reflects the overall quality of the ecosystem of the river basin. Advanced statistical methods are often needed to extract valuable information from the vast amount of data for developing management strategies. Among the measured water quality constituents, total phosphorus is most often the limiting nutrient in freshwater aquatic systems. Relatively low concentrations of phosphorus in surface waters may create eutrophication problems. Phosphorus is a non-conservative constituent. Its time series generally exhibits nonlinear behavior. Linear models are shown to be inadequate. This paper presents a nonlinear state-dependent model for the phosphorous data collected at DeSoto, Kansas. The nonlinear model gives significant reductions in error variance and forecasting error as compared to the best linear autoregressive model identified.     

APPLICATION OF A HYDROLOGIC MODEL FOR LAND USE PLANNING IN FLORIDA1

ABSTRACT: An environmental simulation model of the Upper St. Johns River Basin, Florida, has been developed in order to predict hydrologic responses under proposed management plans. Land use projections for each of 19 hydrologic planning units are provided by a linear programming analysis of agricultural activities. Inputs to the model include rainfall, runoff, evapotranspiration (ET), aquifer properties, topography, soil types, and vegetative patterns. A water balance is developed in the uplands based on infiltration, ET, surface runoff, and groundwater flow. Valley continuity is based on stage-volume relationship for inflows and outflows and a variable roughness coefficient dependent on vegetative patterns. Land use changes form the basis for predicting hydroperiod variation under alternative management schemes. Plans are ranked according to two criteria, deviation from a natural hydroperiod and flood or drought control provided. Results indicate that (1) a single reservoir without irrigation and (2) floodplain preservation plans are superior to (3) multiple reservoir with irrigation and (4) uncontrolled floodplain plans with regard to both criteria.     

NONLINEAR MODELING AND PREDICTION OF A RIVER FLOW SYSTEM1

ABSTRACT: Model estimation and prediction of a river flow system are investigated using nonlinear system identification techniques. We demonstrate how the dynamics of the system, rainfall, and river flow can be modeled using NARMAX (Nonlinear Autoregressive Moving Average with eXogenuous input) models. The parameters of the model are estimated using an orthogonal least squares algorithm with intelligent structure detection. The identification of the nonlinear model is described to represent the relationship between local rainfall and river flow at Enoree station (inputs) and river flow at Whitmire (output) for a river flow system in South Carolina.     

Revisiting a statistical shortcoming when fitting the Langmuir model to sorption data

The Langmuir model is commonly used for describing the sorption behavior of reactive solutes to surfaces and is often fit to sorption data using nonlinear least squares regression. An important assumption of least squares regression is that the predictor variable is error free. In the case of sorption data, this assumption is not valid, and therefore the potential for parameter bias exists. Although alternative regression methods exist that either explicitly account for error in the predictor variable (Model II regression) or minimize the error in the predictor variable, these methods are not commonly used. Therefore, this paper more fully explores the differences in fitted parameters and model fits between these different data fitting methods by fitting P sorption data collected on 26 different soil samples using three different regression methods. For a majority of soils tested in this study, the differences in model fits between the three regression methods were not statistically significant. Statistical differences were observed in over a third of the soils, however, suggesting that errors in the predictor variable may be large enough to produce biased parameter estimates. These results suggest that multiple regression methods should be used when fitting the Langmuir model to sorption data to better assess the potential impact of error on model fits.     

Improvement in HSPF's Low‐Flow Predictions by Implementation of a Power Law Groundwater Storage‐Discharge Relationship

We have enhanced the ability of a widely used watershed model, Hydrologic Simulation Program — FORTRAN (HSPF), to predict low flows by reconfiguring the algorithm that simulates groundwater discharge. During dry weather periods, flow in most streams consists primarily of base flow, that is, groundwater discharged from underlying aquifers. In this study, HSPF's groundwater storage‐discharge relationship is changed from a linear to a more general nonlinear relationship which takes the form of a power law. The nonlinear algorithm is capable of simulating streamflow recession curves that have been found in some studies to better match observed dry weather hydrographs. The altered version of HSPF is implemented in the Chesapeake Bay Program's Phase 5 Model, an HSPF‐based model that simulates nutrient and sediment loads to the Chesapeake Bay, and is tested in the upper Potomac River basin, a 29,950 km² drainage area that is part of the Bay watershed. The nonlinear relationship improved median Nash‐Sutcliffe efficiencies for log daily flows at the model's 45 calibration points. Mean absolute percent error on low‐flow days dropped in five major Potomac River tributaries by up to 12 percentage points, and in the Potomac River itself by four percentage points, where low‐flow days were defined as days when observed flows were in the lowest 5th percentile range. Percent bias on low‐flow days improved by eight percentage points in the Potomac River, from ?11 to ?3%.     

REGIONAL SCALE MODELING OF SURFACE AND GROUND WATER INTERACTION IN THE SNAKE RIVER BASIN1

ABSTRACT: Changes in irrigation and land use may impact discharge of the Snake River Plain aquifer, which is a major contributor to flow of the Snake River in southern Idaho. The Snake River Basin planning and management model (SRBM) has been expanded to include the spatial distribution and temporal attenuation that occurs as aquifer stresses propagate through the aquifer to the river. The SRBM is a network flow model in which aquifer characteristics have been introduced through a matrix of response functions. The response functions were determined by independently simulating the effect of a unit stress in each cell of a finite difference groundwater flow model on six reaches of the Snake River. Cells were aggregated into 20 aquifer zones and average response functions for each river reach were included in the SRBM. This approach links many of the capabilities of surface and ground water flow models. Evaluation of an artificial recharge scenario approximately reproduced estimates made by direct simulation in a ground water flow model. The example demonstrated that the method can produce reasonable results but interpretation of the results can be biased if the simulation period is not of adequate duration.     

OPTIMAL CONTROL OF RESERVOIR RELEASES TO MINIMIZE SEDIMENTATION IN RIVERS AND RESERVOIRS1

ABSTRACT: An optimal control methodology and computational model are developed to evaluate multi‐reservoir release schedules that minimize sediment scour and deposition in rivers and reservoirs. The sedimentation problem is formulated within a discrete‐time optimal control framework in which reservoir releases represent control variables and reservoir bed elevations, storage levels, and river bed elevations represent state variables. Constraints imposed on reservoir storage levels and releases are accommodated using a penalty function method. The optimal control model consists of two interfaced components: a one‐dimensional finite‐difference simulation module used to evaluate flow hydraulics and sediment transport dynamics, and a successive approximation linear quadratic regulator (SALQR) optimization algorithm used to update reservoir release policies and solve the augmented control problem. Hypothetical two‐reservoir and five‐reservoir networks are used to demonstrate the methodology and its capabilities, which is a vital phase towards the development of a more robust optimal control model and application to an existing multiple‐reservoir river network.     

AN EVALUATION OF THE AVAILABLE TECHNIQUES FOR ESTIMATING MISSING FECAL COLIFORM DATA1

ABSTRACT: This paper presents the findings of a study aimed at evaluating the available techniques for estimating missing fecal coliform (FC) data on a temporal basis. The techniques investigated include: linear and nonlinear regression analysis and interpolation functions, and the use of artificial neural networks (ANNs). In all, seven interpolation, two regression, and one ANN model structures were investigated. This paper also investigates the validity of a hypothesis that estimating missing FC data by developing different models using different data corresponding to different dynamics associated with different trends in the FC data may result in a better model performance. The FC data (counts/100 ml) derived from the North Fork of the Kentucky River in Kentucky were employed to calibrate and validate various models. The performance of various models was evaluated using a wide variety of standard statistical measures. The results obtained in this study are able to demonstrate that the ANNs can be preferred over the conventional techniques in estimating missing FC data in a watershed. The regression technique was not found suitable in estimating missing FC data on a temporal basis. Further, it has been found that it is possible to achieve a better model performance by first decomposing the whole data set into different categories corresponding to different dynamics and then developing separate models for separate categories rather than developing a single model for the composite data set.     

Analyzing the Impacts of Dams on Riparian Ecosystems: A Review of Research Strategies and Their Relevance to the Snake River Through Hells Canyon

River damming provides a dominant human impact on river environments worldwide, and while local impacts of reservoir flooding are immediate, subsequent ecological impacts downstream can be extensive. In this article, we assess seven research strategies for analyzing the impacts of dams and river flow regulation on riparian ecosystems. These include spatial comparisons of (1) upstream versus downstream reaches, (2) progressive downstream patterns, or (3) the dammed river versus an adjacent free-flowing or differently regulated river(s). Temporal comparisons consider (4) pre- versus post-dam, or (5) sequential post-dam conditions. However, spatial comparisons are complicated by the fact that dams are not randomly located, and temporal comparisons are commonly limited by sparse historic information. As a result, comparative approaches are often correlative and vulnerable to confounding factors. To complement these analyses, (6) flow or sediment modifications can be implemented to test causal associations. Finally, (7) process-based modeling represents a predictive approach incorporating hydrogeomorphic processes and their biological consequences. In a case study of Hells Canyon, the upstream versus downstream comparison is confounded by a dramatic geomorphic transition. Comparison of the multiple reaches below the dams should be useful, and the comparison of Snake River with the adjacent free-flowing Salmon River may provide the strongest spatial comparison. A pre- versus post-dam comparison would provide the most direct study approach, but pre-dam information is limited to historic reports and archival photographs. We conclude that multiple study approaches are essential to provide confident interpretations of ecological impacts downstream from dams, and propose a comprehensive study for Hells Canyon that integrates multiple research strategies.     

CHAOTIC FORECASTING OF DISCHARGE TIME SERIES: A CASE STUDY1

ABSTRACT: This paper considers the problem of forecasting the discharge time series of a river by means of a chaotic approach. To this aim, we first check for some evidence of chaotic behavior in the dynamic by considering a set of different procedures, namely, the phase portrait of the attractor, the correlation dimension, and the largest Lyapunov exponent. Their joint application seems to confirm the presence of a nonlinear deterministic dynamic of chaotic type. Second, we consider the so‐called nearest neighbors predictor and we compare it with a classical linear model. By comparing these two predictors, it seems that nonlinear river flow modeling, and in particular chaotic modeling, is an effective method to improve predictions.     

如何完善河长制——基于与流域综合管理比较的视角

当前,河长制已成为我国进行河湖管护的综合协调平台,但其面临一系列问题,制约了我国河湖管护效果的进一步提升。世界主要国家的跨部门、跨利益相关方河湖管理协调政策主要是流域综合管理。河长制与流域综合管理都是针对所面临的河湖管理问题,经过长期的演变,逐渐形成的跨区域、跨利益相关方的协调政策,这都是符合所在国家实际情况的。基于公共政策学的基本原理,本文构建了以"政策目标—政策执行者—政策工具"为导向的三维河湖管护协调政策比较框架。通过此框架,分析了河长制存在的主要问题,以及河长制与流域综合管理的主要差异。针对河长制存在的主要问题,基于比较结果提出了相关对策,主要包括:科学设定政策目标,推进多元共治的河湖管护模式,创新运用市场化和信息化政策工具。     

Modeling of power plant impacts on fish populations

A simulation model was developed for assessing the effects of power plant impacts on fish populations. The model was based on biomass changes in a fish population that divided into 4 life stages: eggs, larvae, juveniles, and adults. Metabolism of fish was included in the model as assimilation and respiration with the effects of temperature push-pull regulation. A cohort technique was used for life stage transfers in the young stages. Compensation was included in terms of density-dependent mortality of the young fish. Power plant impacts on a fish population included entrainment (eggs and larvae) and impingement (juveniles and adults).The model was applied to the gizzard shad (Dorosoma cepedianum) population in Pool 14 of the Mississippi River in Illinois and lowa. The nearby nuclear power plant, the Quad-Cities Station, is located on the Illinois side. The simulation of the model with the 1976 field data estimates of the power plant entrainment and impingement predicted a 10 percent potential reduction of the population over 30 years. The simulated reduction of the population with the effects of different river flows showed that the result with the 1976 river flow data gave 1.5 times higher reduction than the results with data of other plant operation years, 1972 through 1975. Because the 1976 data recorded low river flow, the 10 percent reduction quoted above may be high.     

Use of Streamflow Data to Estimate Base Flow/Ground‐Water Recharge For Wisconsin1

Abstract: The average annual base flow/recharge was determined for streamflow‐gaging stations throughout Wisconsin by base‐flow separation. A map of the State was prepared that shows the average annual base flow for the period 1970‐99 for watersheds at 118 gaging stations. Trend analysis was performed on 22 of the 118 streamflow‐gaging stations that had long‐term records, unregulated flow, and provided aerial coverage of the State. The analysis found that a statistically significant increasing trend was occurring for watersheds where the primary land use was agriculture. Most gaging stations where the land cover was forest had no significant trend. A method to estimate the average annual base flow at ungaged sites was developed by multiple‐regression analysis using basin characteristics. The equation with the lowest standard error of estimate, 9.5%, has drainage area, soil infiltration and base flow factor as independent variables. To determine the average annual base flow for smaller watersheds, estimates were made at low‐flow partial‐record stations in 3 of the 12 major river basins in Wisconsin. Regression equations were developed for each of the three major river basins using basin characteristics. Drainage area, soil infiltration, basin storage and base‐flow factor were the independent variables in the regression equations with the lowest standard error of estimate. The standard error of estimate ranged from 17% to 52% for the three river basins.     

Comparison of Measured and MM5 Modeled Meteorology Data for Simulating Flow in a Mountain Watershed1

Herr, Joel W., Krish Vijayaraghavan, and Eladio Knipping, 2010. Comparison of Measured and MM5 Modeled Meteorology Data for Simulating Flow in a Mountain Watershed. Journal of the American Water Resources Association (JAWRA) 46(6):1255–1263. DOI: 10.1111/j.1752-1688.2010.00489.x Abstract: Accurate simulation of time-varying flow in a river system depends on the quality of meteorology inputs. The density of meteorology measurement stations can be insufficient to capture spatial heterogeneity of precipitation, especially in mountainous areas. The Watershed Analysis Risk Management Framework (WARMF) model was applied to the Catawba River watershed of North and South Carolina to simulate flow and water quality in rivers and a series of 11 reservoirs. WARMF was linked with the AMSTERDAM air model to analyze the water quality benefit from reduced atmospheric emissions. The linkage requires accurate simulation of meteorology for all seasons and for all types of precipitation events. WARMF was driven by the mesoscale meteorology model MM5 processed by the Meteorology Chemistry Interface Processor, which provides greater spatial density but less accuracy than meteorology stations. WARMF was also run with measured data from the National Climatic Data Center (NCDC) to compare the performance of the watershed model using measured data vs. modeled meteorology as input. A one year simulation using MM5 modeled meteorology performed better overall than the simulation using NCDC data for the volumetric water balance measure used for calibration, but MM5 represented precipitation from a dissipated hurricane poorly, which propagated into errors of simulated flow.     

A SEPARABLE LINEAR ALGORITHM FOR HYDROPOWER OPTIMIZATION1

ABSTRACT: A deterministic, separable, linear algorithm is presented for maximizing aggregate hydropower production. The method is iterative and amenable to solution using standard LP software. The utility of the technique is demonstrated using several test applications involving a hypothetical single-purpose hydropower reservoir and a monthly increment 20-year flow record from the Gunpowder River in Maryland. The separable linearized forms solved quickly using MPSX on a variety of IBM hardware: 3090-400 VF, 3084 QX, dual processor 4381-3, and an AT/370 personal computer. For comparison purposes, the original nonlinear nonseparable version of the model was also solved using MINOS. This yielded a value of aggregate hydropower marginally higher than that using MPSX. The separable, linearized methodology proved to be a useful and an efficient means of generating good starting points for MINOS. The use of these warm starts effected substantial reductions in MINOS execution times.     

A Framework for Analyzing Longitudinal and Temporal Variation in River Flow and Developing Flow-Ecology Relationships1

Larned, Scott T., David B. Arscott, Jochen Schmidt, and Jan C. Diettrich, 2010. A Framework for Analyzing Longitudinal and Temporal Variation in River Flow and Developing Flow-Ecology Relationships. Journal of the American Water Resources Association (JAWRA) 46(3):541-553. DOI: 10.1111/j.1752-1688.2010.00433.x Abstract: We propose a framework for analyzing longitudinal flow variation and exploring its ecological consequences in four steps: (1) generating longitudinally continuous flow estimates; (2) computing indices that describe site-specific and longitudinal flow variation, including intermittence; (3) quantifying and visualizing longitudinal dynamics; (4) developing quantitative relationships between hydrological indices and ecological variables (flow-ecology relationships). We give examples of each step, using data from a New Zealand river and an empirical longitudinal flow model, ELFMOD. ELFMOD uses spot-gauging data and flow or proxy variable time series to estimate flow magnitude and state (flowing or dry) at user-defined intervals along river sections. Analyses of flow-ecology relationships for the New Zealand river indicated that fish and benthic and hyporheic invertebrate communities responded strongly to variation in mean annual flow permanence, flow duration, dry duration, drying frequency, inter-flood duration, and distances to flowing reaches. To put longitudinal flow variation into a broader context and guide future research, we propose a conceptual model that combines elements of two contrasting perspectives: rivers as longitudinal continua, and rivers as patch mosaics. In this conceptual model, hydrologically complex rivers are composed of linear sequences of nested hydrological gradients, which are bordered by hydrogeomorphic discontinuities, and which collectively generate hydrological dynamics at river-section scales.     

Effect of Human Activities on Overall Trend of Sedimentation in the Lower Yellow River,China

The Yellow River has been intensively affected by human activities, particularly in the past 50 years, including soil–water conservation in the upper and middle drainage basin, flood protection in the lower reaches, and flow regulation and water diversion in the whole drainage basin. All these changes may impact sedimentation process of the lower Yellow River in different ways. Assessing these impacts comprehensively is important for more effective environmental management of the drainage basin. Based on the data of annual river flow, sediment load, and channel sedimentation in the lower Yellow River between 1950 and 1997, the purpose of this paper is to analyze the overall trend of channel sedimentation rate at a time scale of 50 years, and its formative cause. It was found in this study that erosion control measures and water diversion have counteractive impacts on sedimentation rate in the lower Yellow River. Although both annual river flow and sediment decreased, there was no change in channel sedimentation rate. A regression analysis indicated that the sedimentation in the lower Yellow River decreased with the sediment input to the lower Yellow River but increased with the river flow input. In the past 30–40 years, the basin-wide practice of erosion and sediment control measures resulted in a decline in sediment supply to the Yellow River; at the same time, the human development of water resources that required river flow regulation and water diversion caused great reduction in river flow. The former may reduce the sedimentation in the lower Yellow River, but the reduction of river flow increased the sedimentation. When their effects counterbalanced each other, the overall trend of channel sedimentation in the lower Yellow River remained unchanged. This fact may help us to better understand the positive and negative effects of human activities in the Yellow River basin and to pay more attention to the negative effect of the development of water resources. The results of this study demonstrate that, if the overuse of river water cannot be controlled, the reduction of channel sedimentation in the lower Yellow River cannot be realized through the practice of erosion and sediment control measures.     

DETERMINING SIGNIFICANCE AND PRECISION OF ESTIMATED PARAMETERS FOR RUNOFF FROM SEMIARID WATERSHEDS1

ABSTRACT Significant parameters for predicting thunderstorm runoff from small semiarid watersheds are determined using data from the Walnut Gulch watershed in southern Arizona. Based on these data, thunderstorm rainfall is dominant over watershed parameters for predicting runoff from multiple linear regression equations. In some cases antecedent moisture added significantly to the models. A technique is developed for estimating precision of predicted values from multiple linear regression equations. The technique involves matrix methods in estimating the variance of mean predicted values from a regression equation. The estimated variance of the mean predicted value is then used to estimate the variance of an individual predicted value. A computer program is developed to implement these matrix methods and to form confidence limits on predicted values based on both a normality assumption and the Chebyshev inequality.     

A RATING-CURVE METHOD FOR HYDRODYNAMIC SIMULATIONS IN CONTAMINANT TRANSPORT MODELING1

ABSTRACT: Accurate prediction of hydrodynamics is of great importance to modeling contaminant transport and water quality in a river. Flow conditions are needed in estimating potential exposure contamination levels and the recovery time for a no-action alternative in contaminated sediments remediation. Considering highly meandering characteristics of the Buffalo River, New York, a three-dimensional hydrodynamic model was selected to route upstream flows through the 8-km river section with limited existing information based on the model's fully predictive capability and process-oriented feature. The model was employed to simulate changes in water depth and flow velocity with space and time in response to variation in flow rate and/or water surface elevation at boundaries for given bottom morphometry and initial conditions. Flow conditions of the river reach where historical flow data are not available were computed. A rating-curve approach was developed to meet continuous and event contaminant modeling needs. Rating curves (depth-discharge and velocity-discharge relationships) were constructed at selected stations from the 3-D hydrodynamic simulations of individual flow events. The curves were obtained as steady solutions to an unsteady problem. The rating-curve approach serves to link flow information provided by the hydrodynamic model to a contaminant transport model. With the approach, the linking problem resulting from incompatible model dimensions and grid sizes can be solved. The curves will be used to simulate sediment movement and to predict contaminant fate and transport in the river.