Dynamic neuro-fuzzy local modeling system with a nonlinear feature extraction for the online adaptive warning system of river temperature affected by waste cooling water discharge

This paper proposes a dynamic modeling methodology based on a dynamic neuro-fuzzy local modeling system (DNFLMS) with a nonlinear feature extraction technique for an online dynamic modeling task. Prior to model building, a nonlinear feature extraction technique called the Gamma test (GT) is proposed to compute the lowest mean squared error (MSE) that can be achieved and the quantity of data required to obtain a reliable model. Two different DNFLMS modes are developed: (1) an online one-pass clustering and the extended Kalman filtering algorithm (mode 1); and (2) hybrid learning algorithm (mode 2) of extended Kalman filtering algorithm with a back-propagation algorithm trained to the estimated MSE and number of data points determined by a nonlinear feature extraction technique. The proposed modeling methodology is applied to develop an online dynamic prediction system of river temperature to waste cooling water discharge at 1?km downstream from a thermal power station from real-time to time ahead (2?h) sequentially at the new arrival of each item of river, hydrological, meteorological, power station operational data. It is demonstrated that the DNFLMS modes 1 and 2 shows a better prediction performance and less computation time required, compared to a well-known adaptive neural-fuzzy inference system (ANFIS) and a multi-layer perceptron (MLP) trained with the back propagation (BP) learning algorithm, due to local generalization approach and one-pass learning algorithm implemented in the DNFLMS. It is shown that the DNFLMS mode 1 is that it can be used for an online modeling task without a large amount of training set required by the off-line learning algorithm of MLP-BP and ANFIS. The integration of the DNFLMS mode 2 with a nonlinear feature extraction technique shows that it can improve model generalization capability and reduce model development time by eliminating iterative procedures of model construction using a stopping criterion in training and the quantity of required available data in training given by the GT.     

Weather noise impact on the uncertainty of simulated water balance components of river basins

Assessments of hydrological response to climatic changes are characterized by different types of uncertainties. Here, the uncertainty caused by weather noise associated with the chaotic character of atmospheric processes is considered. A technique for estimating such uncertainty in simulated water balance components based on application of the land surface model SWAP and the climate model ECHAM5 is described. The technique is applied for estimating the uncertainties in the simulated water balance components (precipitation, river runoff and evapotranspiration) of some northern river basins of Russia. It is shown that the larger the area of a basin the less the uncertainty. This dependency is smoothed by differences in natural conditions of the basins. Analysis of the spectral densities of water balance components shows that a river basin filters out high-frequency harmonics of spectral density of precipitation (corresponding to synoptic or sub-seasonal scale) during its transformation into evapotranspiration and especially into runoff.

EDITOR D. Koutsoyiannis

ASSOCIATE EDITOR H. Kreibich     

A cellular model of river meandering

River meandering has been successfully modelled using vector based methods, but these can not simulate multiple or braided channels. Conversely, cellular braided river models fail to replicate meandering. This paper describes a new method to simulate river meandering within a cellular model (CAESAR). A novel technique for determining bend radius of curvature on a cell by cell basis is described, that importantly allows regional information on bend curvature to be transferred to local points. This local curvature is then used to drive meandering and lateral erosion through two methods. Key difficulties are identified, including the deposition of material on point bars and cut off development, but the method illustrates how meandering can be integrated within a cellular framework. This demonstrates the potential to produce a single model that can incorporate both meandering and braiding. Copyright © 2006 John Wiley & Sons, Ltd.     

Quantification of the predictive uncertainty of artificial neural network based river flow forecast models

The meaningful quantification of uncertainty in hydrological model outputs is a challenging task since complete knowledge about the hydrologic system is still lacking. Owing to the nonlinearity and complexity associated with the hydrological processes, Artificial neural network (ANN) based models have gained lot of attention for its effectiveness in function approximation characteristics. However, only a few studies have been reported for assessment of uncertainty associated with ANN outputs. This study uses a simple method for quantifying predictive uncertainty of ANN model output through first order Taylor series expansion. The first order partial differential equations of non-linear function approximated by the ANN with respect to weights and biases of the ANN model are derived. A bootstrap technique is employed in estimating the values of the mean and the standard deviation of ANN parameters, and is used to quantify the predictive uncertainty. The method is demonstrated through the case study of Upper White watershed located in the United States. The quantitative assessment of uncertainty is carried out with two measures such as percentage of coverage and average width. In order to show the magnitude of uncertainty in different flow domains, the values are statistically categorized into low-, medium- and high-flow series. The results suggest that the uncertainty bounds of ANN outputs can be effectively quantified using the proposed method. It is observed that the level of uncertainty is directly proportional to the magnitude of the flow and hence varies along time. A comparison of the uncertainty assessment shows that the proposed method effectively quantifies the uncertainty than bootstrap method.     

Spatial prediction of river channel topography by kriging

Topographic information is fundamental to geomorphic inquiry, and spatial prediction of bed elevation from irregular survey data is an important component of many reach‐scale studies. Kriging is a geostatistical technique for obtaining these predictions along with measures of their reliability, and this paper outlines a specialized framework intended for application to river channels. Our modular approach includes an algorithm for transforming the coordinates of data and prediction locations to a channel‐centered coordinate system, several different methods of representing the trend component of topographic variation and search strategies that incorporate geomorphic information to determine which survey data are used to make a prediction at a specific location. For example, a relationship between curvature and the lateral position of maximum depth can be used to include cross‐sectional asymmetry in a two‐dimensional trend surface model, and topographic breaklines can be used to restrict which data are retained in a local neighborhood around each prediction location. Using survey data from a restored gravel‐bed river, we demonstrate how transformation to the channel‐centered coordinate system facilitates interpretation of the variogram, a statistical model of reach‐scale spatial structure used in kriging, and how the choice of a trend model affects the variogram of the residuals from that trend. Similarly, we show how decomposing kriging predictions into their trend and residual components can yield useful information on channel morphology. Cross‐validation analyses involving different data configurations and kriging variants indicate that kriging is quite robust and that survey density is the primary control on the accuracy of bed elevation predictions. The root mean‐square error of these predictions is directly proportional to the spacing between surveyed cross‐sections, even in a reconfigured channel with a relatively simple morphology; sophisticated methods of spatial prediction are no substitute for field data. Copyright © 2007 John Wiley & Sons, Ltd.     

Parameterization of typhoon-induced ocean cooling using temperature equation and machine learning algorithms: an example of typhoon Soulik (2013)

This study proposed three algorithms that can potentially be used to provide sea surface temperature (SST) conditions for typhoon prediction models. Different from traditional data assimilation approaches, which provide prescribed initial/boundary conditions, our proposed algorithms aim to resolve a flow-dependent SST feedback between growing typhoons and oceans in the future time. Two of these algorithms are based on linear temperature equations (TE-based), and the other is based on an innovative technique involving machine learning (ML-based). The algorithms are then implemented into a Weather Research and Forecasting model for the simulation of typhoon to assess their effectiveness, and the results show significant improvement in simulated storm intensities by including ocean cooling feedback. The TE-based algorithm I considers wind-induced ocean vertical mixing and upwelling processes only, and thus obtained a synoptic and relatively smooth sea surface temperature cooling. The TE-based algorithm II incorporates not only typhoon winds but also ocean information, and thus resolves more cooling features. The ML-based algorithm is based on a neural network, consisting of multiple layers of input variables and neurons, and produces the best estimate of the cooling structure, in terms of its amplitude and position. Sensitivity analysis indicated that the typhoon-induced ocean cooling is a nonlinear process involving interactions of multiple atmospheric and oceanic variables. Therefore, with an appropriate selection of input variables and neuron sizes, the ML-based algorithm appears to be more efficient in prognosing the typhoon-induced ocean cooling and in predicting typhoon intensity than those algorithms based on linear regression methods.     

Does an upper limit to river water temperature apply in all places?

There remains continued use of non‐linear, logistic regression models for predicting water temperature from air temperature. A dominant feature of these non‐linear models is an upper bound on river water temperature. This upper bound is often attributed to a large increase in evaporative cooling at high air temperatures, but the exact conditions under which such an increase may occur have not been thoroughly explored. To better understand the appropriateness of the non‐linear model for predicting river water temperatures, it is essential to understand the physical basis for the upper bound and when it should and should not be included in the statistical model. This paper applies and validates an energy balance model against 8 river systems spread across different climate regions of the United States. The energy balance model is then used to develop a diagram relating vapour pressure deficit and air temperature to water temperature. With knowledge of present or future vapour pressure deficit (difference between saturation and actual vapour content in the atmosphere) conditions in a given climate, the diagram can be used to predict the likelihood of an upper bound in the air–water temperature relationship. This investigation offers a fundamental physical explanation of the most appropriate form of statistical models that should be used for predicting future water temperature from air temperature in different geographic regions with different climate conditions. In general, climatic regions that have only a slight increase in vapour pressure deficit with increasing air temperature (typically humid regions) would not be expected to have an upper bound. Conversely, climatic regions in which vapour pressure deficit sharply increases with increasing air temperature (typically arid regions) would be expected to have an upper bound.     

Prediction of daily rainfall state in a river basin using statistical downscaling from GCM output

Conventional statistical downscaling techniques for prediction of multi-site rainfall in a river basin fail to capture the correlation between multiple sites and thus are inadequate to model the variability of rainfall. The present study addresses this problem through representation of the pattern of multi-site rainfall using rainfall state in a river basin. A model based on K-means clustering technique coupled with a supervised data classification technique, namely Classification And Regression Tree (CART), is used for generation of rainfall states from large-scale atmospheric variables in a river basin. The K-means clustering is used to derive the daily rainfall state from the historical daily multi-site rainfall data. The optimum number of clusters in the observed rainfall data is obtained after application of various cluster validity measures to the clustered data. The CART model is then trained to establish relationship between the daily rainfall state of the river basin and the standardized, dimensionally-reduced National Centers for Environmental Prediction/National Center for Atmospheric Research (NCEP/NCAR) reanalysis climatic data set. The relationship thus developed is applied to the General Circulation Model (GCM)-simulated, standardized, bias free large-scale climate variables for prediction of rainfall states in future. Comparisons of the number of days falling under different rainfall states for the observed period and the future give the change expected in the river basin due to global warming. The methodology is tested for the Mahanadi river basin in India.     

Numerical modelling of two-dimensional morphodynamics with applications to river bars and bifurcations

We study the numerical approximation of the two-dimensional morphodynamic model governed by the shallow water and Exner equations to simulate reach-scale two-dimensional morphodynamics of bedload-dominated alluvial rivers. The solution strategy relies on a full coupling of the governing equations within each time step. The resulting system of governing equations contains nonconservative products related to the longitudinal and lateral bed slopes and source terms related to friction. The full problem is solved numerically on unstructured triangular grids, simultaneously updating the principal part and adding the source terms (friction) using a splitting technique.The principal part is solved by means of a novel second-order accurate upwind-biased centred scheme of the finite volume type, while the source terms are added to the problem by solving a system of ordinary differential equations. A new algorithm for treating the wetting-and-drying is also proposed.The model is applied to well-established test problems in order to verify the accuracy of the proposed method, the robustness of the wetting-and-drying algorithm and the ability of the model in dealing with transcritical flows. Finally we test the model ability to reproduce two dimensional morphodynamic processes occurring at the scale of tens of channel widths in bedload dominated alluvial rivers with homogeneous grain size. This is achieved by comparing model outcomes with those of analytical theories and flume experiments on the same morphodynamic processes. These selected “benchmarks” include migrating free bars spontaneously developing in straight reaches, steady bars forced by abrupt river planform changes and the dynamics of channel bifurcations.     

A stochastic conflict resolution model for water quality management in reservoir–river systems

In this paper, optimal operating rules for water quality management in reservoir–river systems are developed using a methodology combining a water quality simulation model and a stochastic GA-based conflict resolution technique. As different decision-makers and stakeholders are involved in the water quality management in reservoir–river systems, a new stochastic form of the Nash bargaining theory is used to resolve the existing conflict of interests related to water supply to different demands, allocated water quality and waste load allocation in downstream river. The expected value of the Nash product is considered as the objective function of the model which can incorporate the inherent uncertainty of reservoir inflow. A water quality simulation model is also developed to simulate the thermal stratification cycle in the reservoir, the quality of releases from different outlets as well as the temporal and spatial variation of the pollutants in the downstream river. In this study, a Varying Chromosome Length Genetic Algorithm (VLGA), which has computational advantages comparing to other alternative models, is used. VLGA provides a good initial solution for Simple Genetic Algorithms and comparing to Stochastic Dynamic Programming (SDP) reduces the number of state transitions checked in each stage. The proposed model, which is called Stochastic Varying Chromosome Length Genetic Algorithm with water Quality constraints (SVLGAQ), is applied to the Ghomrud Reservoir–River system in the central part of Iran. The results show, the proposed model for reservoir operation and waste load allocation can reduce the salinity of the allocated water demands as well as the salinity build-up in the reservoir.     

Spectral variations of underwater river sounds

Passive acoustic monitoring of the self‐generated noise of particle impacts has been shown to be correlated to bedload flux and bedload size. However, few studies have concentrated on the role of acoustic wave propagation in a river. For the first time, the river environment is modeled as a Pekeris waveguide, where a wave number integration technique is used to predict the transformation of sounds through their propagation paths. Focusing on the distance of a hydrophone from the channel bed and cutting off the low frequencies produced by impacts between gravel particles, we demonstrate that acoustic propagation modifies the spectral content of bedload‐generated sound. Acoustic signals analyzed with the proposed model are interpreted by comparison to Helley–Smith bedload data obtained during flood conditions on the large gravel‐bedded Arc‐en‐Maurienne River, France. This study shows that careful attention to acoustic propagation effects is required when estimating bedload grain size distribution with hydrophones in rivers, especially for rivers with slopes higher than 1%. Bedload monitoring with a hydrophone is particularly appropriate for large gravel‐bed rivers – especially so during large floods, when in situ sampling is difficult or impractical and the impact of acoustic propagation is weaker relative to the self‐generated noise of bedload impacts. Copyright © 2017 John Wiley & Sons, Ltd.     

The influence of storm characteristics and catchment conditions on extreme flood response : A case study based on the brue river basin,U.K.

Methods for estimating the magnitude of extreme floods are reviewed. A method which combines a probabilistic storm transposition technique with a physically-based distributed rainfallrunoff model is described. Synthetic storms with detailed spatial and temporal distributions are generated and applied to the calibrated model of the Brue river basin, U.K. (area 135 km²). The variability of catchment response due to storm characteristics (storm area, storm duration, storm movement, storm shape and within storm variation) and initial catchment wetness conditions is investigated. A probabilistic approach to estimating the return periods of extreme catchment responses is suggested.     

Simplified nonlinear modeling of river flow routing

Management of open-channel flow systems requires accurate models of flow transfer. This article presents a simple nonlinear model representative of the flow transfer in a river reach. The model is obtained through linearization of a physical model, simplification using the cumulant matching method and analytic identification of a nonlinear model coinciding with the linear model around equilibrium points, corresponding to the hydraulic permanent regimes. The methodology is illustrated on the diffusive wave equation and the Saint-Venant equations. The obtained nonlinear models are compared in simulation to the initial models. The nonlinear model is shown to ensure mass conservation, despite the variable delay element of the model. The proposed model can reproduce the nonlinear behavior of the time-delay with discharge variations. It is well-suited for fast simulations, flow forecasting, and for controller design.     

Empirical and conceptual modelling of the suspended sediment dynamics in a large tropical African river: the Upper Niger river basin

A 7-year sediment transport monitoring on the Upper Niger rivers was used to study the relationship between suspended sediment concentration and river discharge. During annual floods, these relationships show positive hysteresis. This paper presents the results of two models that estimate the time evolution of suspended sediment concentration using water discharge data only. The first model is based on a statistical approach using two relationships, one for the rising stage period of the flood and one for the recession period of the annual flood; the second model is a lumped conceptual one; it supposes that the sediment flux observed in the river comes from two different sources of sediment and that these two sources may be regarded as two different reservoirs. The erosion of the first reservoir represents hillslope erosion observed during the runoff season. Sediment supply from this ‘reservoir’ is limited in time because depletion occurs during the runoff season. The second reservoir is unlimited in time and quantity and its erosion represents contributions coming from bank erosion and mobilisation of deposits in the channel network.

Both of the models are compared with a simple rating curve based model. The model results show that the conceptual model has the highest efficiency to reproduce from weekly discharge only the time evolution of weekly suspended sediment concentrations, the time evolution of weekly sediment fluxes, and the global annual sediment yields.     

Visualization of the turbulent flow structure in a gravel-bed river

A novel technique for visualizing turbulent flow data from a gravel-bed river is presented. The time development of flow velocity and shear stress at three heights is displayed using a computer program. This can be used to observe how the fluctuations of velocity and shear stress interact both spatially and temporally. We highlight examples of flow events which are important for the understanding of flow dynamics. The visualization suggests that the turbulent flow-field is characterized not only by coherence over time at a point, but also by spatial interdependence between points. We suggest that this new visualization approach will assist further interpretation of statistical analyses of turbulent signals, as well as focusing future measurement strategies by providing a clearer spatio-temporal picture of the flow structure.     

Cooling rates of group IVA iron meteorites determined from a ternary Fe-Ni-P model

Cooling rates have been determined for twelve group IVA iron meteorites using a ternary (Fe-Ni-P) model that simulates the growth of the Widmanstätten pattern. The new ternary model is governed by a set of differential diffusion equations that are coupled through the phase growth velocity and elemental concentration profiles. Measured ternary diffusivities and phase diagram solubilities were extrapolated below 500°C for use in the model. The model is more sophisticated than previous ones in that P as well as Ni gradients are calculated, ternary α-γ tie lines are allowed to vary, and ternary diffusivities are used.Output from the simulation is used to create a family of cooling rate curves on plots of central taenite Ni vs. log taenite half-width for each meteorite. A comparison of measured data to the cooling rate curves yields unique meteorite cooling rates. The measured cooling rates for the twelve IVA irons vary inversely with Ni content by over an order of magnitude (4–200°C/Myr). It is proposed that the group IVA irons were accommodated at various depths in an asteroidal-sized body.     

Cooling,degassing and compaction of rhyolitic ash flow tuffs: a computational model

Previous models of degassing, cooling and compaction of rhyolitic ash flow deposits are combined in a single computational model that runs on a personal computer. The model applies to a broader range of initial and boundary conditions than Riehle's earlier model, which did not integrate heat and mass flux with compaction and which for compound units was limited to two deposits. Model temperatures and gas pressures compare well with simple measured examples. The results indicate that degassing of volatiles present at deposition occurs within days to a few weeks. Compaction occurs for weeks to two to three years unless halted by devitrification; near-emplacement temperatures can persist for tens of years in the interiors of thick deposits. Even modest rainfall significantly chills the upper parts of ash deposits, but compaction in simple cooling units ends before chilling by rainwater influences cooling of the interior of the sheet. Rainfall does, however, affect compaction at the boundaries of deposits in compound cooling units, because the influx of heat from the overlying unit is inadequate to overcome heat previously lost to vaporization of water. Three density profiles from the Matahina Ignimbrite, a compound cooling unit, are fairly well reproduced by the model despite complexities arising from numerous cooling breaks. Uncertainties in attempts to correlate in detail among the profiles may be the result of the non-uniform distribution of individual deposits. Regardless, it is inferred that model compaction is approximately valid. Thus the model should be of use in reconstructing the emplacement history of compound ash deposits, for inferring the depositional environments of ancient deposits and for assessing how long deposits of modern ash flows are capable of generating phreatic eruptions or secondary ash flows.     

Data preprocessing for river flow forecasting using neural networks: Wavelet transforms and data partitioning

The evaluation of surface water resources is a necessary input to solving water management problems. Neural network models have been trained to predict monthly runoff for the Tirso basin, located in Sardinia (Italy) at the S. Chiara section. Monthly time series data were available for 69 years and are characterized by non-stationarity and seasonal irregularity, which is typical of a Mediterranean weather regime. This paper investigates the effects of data preprocessing on model performance using continuous and discrete wavelet transforms and data partitioning. The results showed that networks trained with pre-processed data performed better than networks trained on undecomposed, noisy raw signals. In particular, the best results were obtained using the data partitioning technique.     

Measuring river velocities using GPS River Flow Tracers (GRiFTers)

Existing methods of measuring flow velocities in natural rivers are largely based on series of point measurements. Acquisition of these data can be time consuming and difficult, especially in high flow conditions. This paper introduces the use of GPS drifters (termed GRiFTers) to measure surface flow velocities in a 400 m reach of the River Swale, UK. Over 10 000 measurements were made in a 3 hour period and aggregated over a 2 m grid to generate a genuine distributed representation of flow across the reach. The technique shows great promise to provide new insights into flow patterns over long reaches of rivers, over a range of flow conditions, and may also provide valuable data for numerical model validation. Copyright © 2007 John Wiley & Sons, Ltd.     

Disturbed periodic model for river meanders

Although river meanders are not perfectly regular their serial statistics show periodic tendencies which cannot be explained by previous stochastic models. The regular and random approaches to meander geometry can be reconciled in a disturbed periodic model with separate scale, sinuosity, and irregularity parameters. Meandering is viewed as a deterministic oscillation but irregularity is introduced by quasi-random variability in valley-floor topography and materials. For stability such a model needs either a Bagnold type limit on bend curvature or frictional damping of the oscillatory response to individual disturbances. Realistic statistical properties are derived for the second case. The differential equation for direction can be approximated by a second-order autoregression, which generates realistic simulated patterns and gives a good fit to natural direction series.