Modelling daily suspended sediment of rivers in Turkey using several data-driven techniques / Modélisation de la charge journalière en matières en suspension dans des rivières turques à l'aide de plusieurs techniques empiriques

Abstract

The transport of sediment load in rivers is important with respect to pollution, channel navigability, reservoir filling, longevity of hydroelectric equipment, fish habitat, river aesthetics and scientific interest. However, conventional sediment rating curves cannot estimate sediment load accurately. An adaptive neuro-fuzzy technique is investigated for its ability to improve the accuracy of the streamflow—suspended sediment rating curve for daily suspended sediment estimation. The daily streamflow and suspended sediment data for four stations in the Black Sea region of Turkey are used as case studies. A comparison is made between the estimates provided by the neuro-fuzzy model and those of the following models: radial basis neural network (RBNN), feed-forward neural network (FFNN), generalized regression neural network (GRNN), multi-linear regression (MLR) and sediment rating curve (SRC). Comparison of results reveals that the neuro-fuzzy model, in general, gives better estimates than the other techniques. Among the neural network techniques, the RBNN is found to perform better than the FFNN and GRNN.     

Predicting longitudinal dispersion coefficient in natural streams by artificial intelligence methods

In this study, three artificial neural network methods, i.e. feed forward back propagation, the radial basis function neural network, and the generalized regression neural network are employed to compute the longitudinal dispersion coefficient in order to evaluate its behaviour in predicting dispersion characteristics in natural streams. These methods, which use hydraulic and geometrical data to predict dispersion coefficients, can easily be applied to natural streams and are proven to be superior in explaining their dispersion characteristics more precisely than existing equations. This method of predicting the longitudinal dispersion coefficient in river flows was tested on 65 data sets, obtained by researchers from 30 rivers in the USA. Results using the models are compared with results obtained in many other studies, and are shown to be more accurate than the other methods considered. Copyright © 2008 John Wiley & Sons, Ltd.     

Application of a neural network model in establishing a stage–discharge relationship for a tidal river

This paper presents the applicability of neural network (NN) modelling for forecasting and filtering problems. The multilayer feedforward (MLFF) network was first constructed to forecast the tidal‐level variations at the mouth of the River Chao Phraya in Thailand. Unlike the well‐known conventional harmonic analysis, the NN model uses a set of previous data for learning and then forecasting directly the time‐series of tidal levels. It was found that lead time of 1 to 24 hourly tidal levels can be predicted successfully using only a short‐time hourly learning data. The MLFF network was further used to establish a stage–discharge relationship for the tidal river. The results show a considerably better performance of the NN model over the conventional models. In addition, the stage–discharge relationship obtained by the NN model can indicate reasonably well the important behaviour of the tidal influences. Copyright © 2003 John Wiley & Sons, Ltd.     

Daily pan evaporation modelling using multi‐layer perceptrons and radial basis neural networks

This paper reports on investigations of the abilities of three different artificial neural network (ANN) techniques, multi‐layer perceptrons (MLP), radial basis neural networks (RBNN) and generalized regression neural networks (GRNN) to estimate daily pan evaporation. Different MLP models comprising various combinations of daily climatic variables, that is, air temperature, solar radiation, wind speed, pressure and humidity were developed to evaluate the effect of each of these variables on pan evaporation. The MLP estimates are compared with those of the RBNN and GRNN techniques. The Stephens‐Stewart (SS) method is also considered for the comparison. The performances of the models are evaluated using root mean square errors (RMSE), mean absolute error (MAE) and determination coefficient (R²) statistics. Based on the comparisons, it was found that the MLP and RBNN computing techniques could be employed successfully to model the evaporation process using the available climatic data. The GRNN was found to perform better than the SS method. Copyright © 2008 John Wiley & Sons, Ltd.     

Simulating and predicting river discharge time series using a wavelet‐neural network hybrid modelling approach

Accurate simulation and prediction of the dynamic behaviour of a river discharge over any time interval is essential for good watershed management. It is difficult to capture the high‐frequency characteristics of a river discharge using traditional time series linear and nonlinear model approaches. Therefore, this study developed a wavelet‐neural network (WNN) hybrid modelling approach for the predication of river discharge using monthly time series data. A discrete wavelet multiresolution method was employed to decompose the time series data of river discharge into sub‐series with low (approximation) and high (details) frequency, and these sub‐series were then used as input data for the artificial neural network (ANN). WNN models with different wavelet decomposition levels were employed to predict river discharge 48 months ahead of time. Comparison of results from the WNN models with those of the ANN models alone indicated that WNN models performed a more accurate prediction. Copyright © 2011 John Wiley & Sons, Ltd.     

A comparative study of three neural network forecast combination methods for simulated river flows of different rainfall—runoff models

Abstract

The performances of three artificial neural network (NN) methods for combining simulated river flows, based on three different neural network structures, are compared. These network structures are: the simple neural network (SNN), the radial basis function neural network (RBFNN) and the multi-layer perceptron neural network (MLPNN). Daily data of eight catchments, located in different parts of the world, and having different hydrological and climatic conditions, are used to enable comparisons of the performances of these three methods to be made. In the case of each catchment, each neural network combination method synchronously uses the simulated river flows of four rainfall—runoff models operating in design non-updating mode to produce the combined river flows. Two of these four models are black-box, the other two being conceptual models. The results of the study show that the performances of all three combination methods are, on average, better than that of the best individual rainfall—runoff model utilized in the combination, i.e. that the combination concept works. In terms of the Nash-Sutcliffe model efficiency index, the MLPNN combination method generally performs better than the other two combination methods tested. For most of the catchments, the differences in the efficiency index values of the SNN and the RBFNN combination methods are not significant but, on average, the SNN form performs marginally better than the more complex RBFNN alternative. Based on the results obtained for the three NN combination methods, the use of the multi-layer perceptron neural network (MLPNN) is recommended as the appropriate NN form for use in the context of combining simulated river flows.     

Stage‐discharge uncertainty derived with a non‐stationary rating curve in the Choluteca River,Honduras

Uncertainty in discharge data must be critically assessed before data can be used in, e.g. water resources estimation or hydrological modelling. In the alluvial Choluteca River in Honduras, the river‐bed characteristics change over time as fill, scour and other processes occur in the channel, leading to a non‐stationary stage‐discharge relationship and difficulties in deriving consistent rating curves. Few studies have investigated the uncertainties related to non‐stationarity in the stage‐discharge relationship. We calculated discharge and the associated uncertainty with a weighted fuzzy regression of rating curves applied within a moving time window, based on estimated uncertainties in the observed rating data. An 18‐year‐long dataset with unusually frequent ratings (1268 in total) was the basis of this study. A large temporal variability in the stage‐discharge relationship was found especially for low flows. The time‐variable rating curve resulted in discharge estimate differences of ? 60 to + 90% for low flows and ± 20% for medium to high flows when compared to a constant rating curve. The final estimated uncertainty in discharge was substantial and the uncertainty limits varied between ? 43 to + 73% of the best discharge estimate. Copyright © 2010 John Wiley & Sons, Ltd.     

An analytical method of stage–fall–discharge rating

This paper presents an analytical method for establishing a stage–fall–discharge rating using hydraulic performance graphs (HPG). The rating curves derived from the HPG are used as the basis to establish the functional relation of stage, fall and discharge through regression analysis following the USGS procedure. In doing so, the conventional trial‐and‐error process can be avoided and the associated uncertainties involved may be reduced. For illustration, the proposed analytical method is applied to establish stage–fall–discharge relations for the Keelung River in northern Taiwan to examine its accuracy and applicability in an actual river. Based on the data extracted from the HPG for the Keelung River, one can establish a stage–fall–discharge relation that is more accurate than the one obtained by the conventionally used relation. Furthermore, the discharges obtained from the proposed rating method are verified through backwater analysis for measured high water level events. The results indicate that the analytical stage–fall–discharge rating method is capable of circumventing the shortcomings of those based on single‐station data and, consequently, enhancing the reliability of flood estimation and forecasting. Copyright © 2008 John Wiley & Sons, Ltd.     

Regionalizing mean annual flow and daily flow variability for basin‐scale sediment and nutrient modelling

River discharges vary strongly through time and space, and quantifying this variability is fundamental to understanding and modelling river processes. The river basin is increasingly being used as the unit for natural resource planning and management; to facilitate this, basin‐scale models of material supply and transport are being developed. For many basin‐scale planning activities, detailed rainfall‐runoff modelling is neither necessary nor tractable, and models that capture spatial patterns of material supply and transport averaged over decades are sufficient. Nevertheless, the data to describe the spatial variability of river discharge across large basins for use in such models are often limited, and hence models to predict river discharge at the basin scale are required. We describe models for predicting mean annual flow and a non‐dimensional measure of daily flow variability for every river reach within a drainage network. The models use sparse river gauging data, modelled grid surfaces of mean annual rainfall and mean annual potential evapotranspiration, and a network accumulation algorithm. We demonstrate the parameterization and application of the models using data for the Murrumbidgee basin, in southeast Australia, and describe the use of these predictions in modelling sediment transport through the river network. The regionalizations described contain less uncertainty, and are more sensitive to observed spatial variations in runoff, than regionalizations based on catchment area and rainfall alone. Copyright © 2006 John Wiley & Sons, Ltd.     

Prediction of flow rate of karstic springs using support vector machines

Complex void space structure and flow patterns in karstic aquifers render behaviour prediction of karstic springs difficult. Four support vector regression-based models are proposed to predict flow rates from two adjacent karstic springs in Greece (Mai Vryssi and Pera Vryssi). Having no accurate estimates of the groundwater flow pattern, we used four kernels: linear, polynomial, Gaussian radial basis function and exponential radial basis function (ERBF). The data used for training and testing included daily and mean monthly precipitation, and spring flow rates. The support vector machine (SVM) performance depends on hyper-parameters, which were optimized using a grid search approach. Model performance was evaluated using root mean square error and correlation coefficient. Polynomial kernel performed better for Mai Vryssi and the ERBF for Pera Vryssi. All models except one performed better for Pera Vryssi. Our models performed better than generalized regression neural network, radial basis function neural network and ARIMA models.     

The potential of different ANN techniques in evapotranspiration modelling

The potential of three different artificial neural network (ANN) techniques, the multi‐layer perceptrons (MLPs), radial basis neural networks (RBNNs) and generalized regression neural networks (GRNNs), in modelling of reference evapotranspiration (ET₀) is investigated in this paper. Various daily climatic data, that is, solar radiation, air temperature, relative humidity and wind speed from two stations, Pomona and Santa Monica, in Los Angeles, USA, are used as inputs to the ANN techniques so as to estimate ET₀ obtained using the FAO‐56 Penman–Monteith (PM) equation. In the first part of the study, a comparison is made between the estimates provided by the MLP, RBNN and GRNN and those of the following empirical models: The California Irrigation Management Information System (CIMIS) Penman (1985), Hargreaves (1985) and Ritchie (1990). In this part of the study, the empirical models are calibrated using the standard FAO‐56 PM ET₀ values. The estimates of the ANN techniques are also compared with those of the calibrated empirical models. Mean square errors, mean absolute errors and determination coefficient statistics are used as comparing criteria for the evaluation of the models' performances. Based on the comparisons, it is found that the MLP and RBNN techniques could be employed successfully in modelling the ET₀ process. In the second part of the study, the potential of ANN techniques and the empirical methods in ET₀ estimation using nearby station data is investigated. Among the models, the calibrated Hargreaves model is found to perform better than the others. Copyright © 2007 John Wiley & Sons, Ltd.     

Estimation of river flow by artificial neural networks and identification of input vectors susceptible to producing unreliable flow estimates

Reliable river flow estimates are crucial for appropriate water resources planning and management. River flow forecasting can be conducted by conceptual or physical models, or data-driven black box models. Development of physically-based models requires an understanding of all the physical processes which impact a natural process and the interactions among them. Since identification of the relationships among these physical processes is very difficult, data-driven approaches have recently been utilized in hydrological modeling. Artificial neural networks are one of the widely used data-driven approaches for modeling hydrological processes. In this study, estimation of future monthly river flows for Guvenc River, Ankara is conducted using various artificial neural network models. Success of artificial neural network models relies on the availability of adequate data sets. A direct mapping from inputs to outputs without consideration of the complex relationships among the dependent and independent variables of the hydrological process is identified. In this study, past precipitation, river flow data, and the associated month are used to predict future river flows for Guvenc River. Impacts of various input patterns, number of training cycles, and initial values assigned to the weights of the connections are investigated. One of the major weaknesses of artificial neural networks is that they may fail to generate good estimates for extreme events, i.e. events that do not occur at all or often enough in the training data set. It is very important to be able to identify such unlikely events. A fuzzy c-means algorithm is used in this study to cluster the training and validation input vectors into regular and extreme events so that the user will have an idea about the risk of the artificial neural network model to generate unreliable results.     

Development of spatial regression models for predicting summer river temperatures from landscape characteristics: Implications for land and fisheries management

There is increasing demand for models that can accurately predict river temperature at the large spatial scales appropriate to river management. This paper combined summer water temperature data from a strategically designed, quality controlled network of 25 sites, with recently developed flexible spatial regression models, to understand and predict river temperature across a 3,000 km² river catchment. Minimum, mean and maximum temperatures were modelled as a function of nine potential landscape covariates that represented proxies for heat and water exchange processes. Generalised additive models were used to allow for flexible responses. Spatial structure in the river network data (local spatial variation) was accounted for by including river network smoothers. Minimum and mean temperatures decreased with increasing elevation, riparian woodland and channel gradient. Maximum temperatures increased with channel width. There was greater between‐river and between‐reach variability in all temperature metrics in lower‐order rivers indicating that increased monitoring effort should be focussed at these smaller scales. The combination of strategic network design and recently developed spatial statistical approaches employed in this study have not been used in previous studies of river temperature. The resulting catchment scale temperature models provide a valuable quantitative tool for understanding and predicting river temperature variability at the catchment scales relevant to land use planning and fisheries management and provide a template for future studies.     

Using satellite altimetry data to augment flow estimation techniques on the Mekong River

Satellite altimetry is routinely used to provide levels for oceans or large inland water bodies from space. By utilizing retracking schemes specially designed for inland waters, meaningful river stages can also be recovered when standard techniques fail. Utilizing retracked waveforms from ERS‐2 and ENVISAT along the Mekong, comparisons against observed stage measurements show that the altimetric measurements have a root mean square error (RMSE) of 0·44–0·65 m for ENVISAT and 0·46–0·76 m for ERS‐2. For many applications, however, stage is insufficient because discharge is the primary requirement. Investigations were therefore undertaken to estimate discharges at a downstream site (Nakhon Phanom (NP)) assuming that in situ data are available at a site 400 km upstream (Vientiane). Two hypothetical, but realistic scenarios were considered. Firstly, that NP was the site of a de‐commissioned gauge and secondly, that the site has never been gauged. Using both scenarios, predictions were made for the daily discharge using methods with and without altimetric stage data. In the first scenario using a linear regression approach the altimetry data improved the Nash‐Sutcliffe r² value from 0·884 to 0·935. The second scenario used known river cross‐sections while lateral inflows were inferred from a hydrological model: this scenario gave an increase in the r² value from 0·823 to 0·893. The use of altimetric stage data is shown to improve estimated discharges and further applications are discussed. Copyright © 2010 John Wiley & Sons, Ltd.     

How to evaluate the quality of coarse‐resolution DEM‐derived drainage networks

The quality of digital elevation model (DEM)‐derived river drainage networks (RDNs) is influenced by DEM quality, basin physical characteristics, scale, and algorithms used; these factors should not be neglected. However, few research studies analyse the different evaluation approaches used in the literature with respect to adequacy, meaning of the results, advantages, and limitations. Focusing on coarse‐resolution networks, this paper reviews the use of these techniques and offers new insights on these issues. Additionally, we propose adaptations for selected metrics and discuss distinct interpretations for the evaluation of RDNs derived at different spatial resolutions (1, 5, 10, 20, and 30 km) considering the Uruguay River basin (206,000 km²) as a case study. The results demonstrate that lumped basin/river characteristics and basin delineation analysis should not be used as evaluation criteria for RDN quality; however, some of these metrics offer useful complementary information. Percentage of the DEM‐derived RDN within a uniform buffer placed around a river network considered as reference and mean separation distance between these two networks are more suitable metrics, but the former is insensitive to serious errors. The change in reference from a fine‐scale network to a coarse‐resolution manual tracing network significantly augments the discrepancy of these largest errors when the mean distance metric was applied, and visual comparison analysis is necessary to interpret the results for other metrics. We recommend the use of the mean distance metric in combination with a detailed visual assessment, the importance of which increases as the resolution coarsens. In both cases, the impact of network quality can be further refined by quantifying the basin shape and river length errors.     

Defining and measuring braiding intensity

Geomorphological studies of braided rivers still lack a consistent measurement of the complexity of the braided pattern. Several simple indices have been proposed and two (channel count and total sinuosity) are the most commonly applied. For none of these indices has there been an assessment of the sampling requirements and there has been no systematic study of the equivalence of the indices to each other and their sensitivity to river stage. Resolution of these issues is essential for progress in studies of braided morphology and dynamics at the scale of the channel network. A series of experiments was run using small‐scale physical models of braided rivers in a 3 m ∞ 20 m flume. Sampling criteria for braid indices and their comparability were assessed using constant‐discharge experiments. Sample hydrographs were run to assess the effect of flow variability. Reach lengths of at least 10 times the average wetted width are needed to measure braid indices with precision of the order of 20% of the mean. Inherent variability in channel pattern makes it difficult to achieve greater precision. Channel count indices need a minimum of 10 cross‐sections spaced no further apart than the average wetted width of the river. Several of the braid indices, including total sinuosity, give very similar numerical values but they differ substantially from channel‐count index values. Consequently, functional relationships between channel pattern and, for example, discharge, are sensitive to the choice of braid index. Braid indices are sensitive to river stage and the highest values typically occur below peak flows of a diurnal (melt‐water) hydrograph in pro‐glacial rivers. There is no general relationship with stage that would allow data from rivers at different relative stage to be compared. At present, channel count indices give the best combination of rapid measurement, precision, and range of sources from which measurements can be reliably made. They can also be related directly to bar theory for braided pattern development. Copyright © 2008 John Wiley & Sons, Ltd.     

Developing channel and floodplain dimensions with limited data: a case study in the Amazon Basin

This research builds on the concept of hydraulic geometry and presents a methodology for estimating bankfull discharge and the hydraulic geometry coefficients and exponents for a station using limited data; only stage‐discharge and Landsat imagery. The approach is implemented using 82 streamflow gauging locations in the Amazon Basin. Using the estimated values for the hydraulic geometry relations, bankfull discharge, discharge data above bankfull and upstream drainage area at each site, relationships for estimating channel and floodplain characteristics as a function of drainage area are developed. Specifically, this research provides relationships for estimating bankfull discharge, bankfull depth, bankfull width, and floodplain width as a function of upstream drainage area in the Amazon Basin intended for providing reasonable cross‐section estimates for large scale hydraulic routing models. The derived relationships are also combined with a high resolution drainage network to develop relationships for estimating cumulative upstream channel lengths and surface areas as a function of the specified minimum channel width ranging from 2 m to 1 km (i.e. threshold drainage areas ranging from 1 to 431,000 km²). At the finest resolution (i.e. all channels greater than 2 m or a threshold area of 1 km²), the Amazon Basin contains approximately 4.4 million kilometers of channels with a combined surface area of 59,700 km². The intended use of these relationships is for partitioning total floodable area (channels versus lakes and floodplain lakes) obtained from remote sensing for biogeochemical applications (e.g. quantifying CO₂ evasion in the Amazon Basin). Copyright © 2011 John Wiley & Sons, Ltd.     

Increasing risk and uncertainty of flooding in the Mississippi River basin

Flooding in the Mississippi basin has become increasingly uncertain, and a succession of progressively higher, peak annual water levels is observed at many sites. Many record levels set in the central USA by the huge 1993 flood have already been superseded. Methodology developed elsewhere that recognizes trends of river stages is used to estimate present‐day flood risk at 27 sites in the Mississippi basin that have >100 years of continuous stage record. Unlike official estimates that are fundamentally based on discharge, this methodology requires only data on river stage. A novel plot linearizes the official flood levels that are indirectly derived from the complex, discharge‐based calculations and demonstrates that the neglect of trends has resulted in the effective use of undersized means and standard deviations in flood risk analysis. A severe consequence is that official “base flood” levels are underestimated by 0.4 to 2 m at many sites in the central USA.     

Applying Neural Network Classification to Obtain Mangrove Landscape Characteristics for Monitoring the Travel Environment Quality on the Beihai Coast of Guangxi,P. R. China

The spectral characteristics of mangroves on the Beihai Coast of Guangxi, P. R. China are acquired on the basis of spectral data from field measurements. Following this, the 3‐layer reverse‐conversing neural networks (NN) classification technology is used to analyze the Landsat TM5 image obtained on January 8, 2003. It is detailed enough to facilitate the introduction of the algorithm principle and trains project of the neural network. Neural network algorithms have characteristics including large‐scale data handling and distributing information storage. This research firstly analyzes the necessity and complexity of this translation system, and then introduces the strong points of the neural network. Processing mangrove landscape characteristics by using neural network is an important innovation, with great theoretical and practical significance. This kind of neural network can greatly improve the classification accuracy. The spatial resolution of Landsat TM5 is high enough to facilitate the research, and the false color composite from 3‐, 4‐, and 5‐bands has a clear boundary and provides a significant quantity of information and effective images. On the basis of a field survey, the exported layers are defined as mangrove, vegetation, bare land, wetlands and shrimp pool. TM satellite images are applied to false color composites by using 3‐, 4‐, and 5‐bands, and then a supervised classification model is used to classify the image. The processing method of hyper‐spectrum remote sensing allows the spectral characteristics of the mangrove to be determined, and integrates the result with the NN classification for the false color composite by using 3‐, 4‐, and 5‐bands. The network model consists of three layers, i. e., the input layer, the hidden layer, and the output layer. The input layer number of classification is defined as 3, and the hidden layers are defined as 5 according to the function operation. The control threshold is 0.9. The training ratio is 0.2. The maximum permit error is 0.08. The classification precision reaches 86.86%. This is higher than the precision of maximal parallel classification (50.79%) and the spectrum angle classification (75.39%). The results include the uniformity ratio (1.7789), the assembly ratio (0.6854), the dominance ratio (–1.5850), and the fragmentation ratio (0.0325).