scholarly journals Hindcasting and Forecasting Total Suspended Sediment Concentrations Using a NARX Neural Network

2021 ◽  
Vol 13 (1) ◽  
pp. 363
Author(s):  
Vladimir J. Alarcon
Keyword(s):  
Neural Network ◽  
Suspended Sediment ◽  
Urban Areas ◽  
Input Data ◽  
Suspended Sediments ◽  
Efficiency Index ◽  
Open Loop ◽  
Model Parameters ◽  
Creek Watershed ◽  
Suspended Sediment Concentrations

Estimating and forecasting suspended sediments concentrations in streams constitutes a valuable asset for sustainable land management. This research presents the development of a non-linear autoregressive exogenous neural network (NARX) for forecasting sediment concentrations at the exit of Francia Creek watershed (Valparaiso, Chile). Details are presented on input data selection, data splitting, selection of model architecture, determination of model structure, NARX training (optimization of model parameters), and model validation (hindcasting and forecasting). The study explored if the developed artificial neural network model is valid for forecasting daily suspended sediment concentrations for a complete year, capturing seasonal trends, and maximum and baseflow concentrations. Francia Creek watershed covers approximately 3.24 km2. Land cover within the catchment consists mainly of native and exotic vegetation, eroded soil, and urban areas. Input data consisting of precipitation and stream flow time-series were fed to a NARX network for forecasting daily suspended sediments (SST) concentrations for years 2013–2014, and hindcasting for years 2008–2010. Training of the network was performed with daily SST, precipitation, and flow data from years 2012 and 2013. The resulting NARX net consisted of an open-loop, 12-node hidden layer, 100 iterations, using Bayesian regularization backpropagation. Hindcasting of daily and monthly SST concentrations for years 2008 through 2010 was successful. Daily SST concentrations for years 2013 and 2014 were forecasted successfully for baseflow conditions (R2 = 0.73, NS = 0.71, and Kling-Gupta efficiency index (K-G) = 0.84). Forecasting daily SST concentrations for year 2014 was within acceptable statistical fit and error margins (R2 = 0.53, NS = 0.47, K-G = 0.60, d = 0.82). Forecasting of monthly maximum SST concentrations for the two-year period (2013 and 2014) was also successful (R2 = 0.69, NS = 0.60, K-G = 0.54, d = 0.84).

scholarly journals Curvas-chave de descargas de sedimentos em suspensão no Baixo São Francisco

2020 ◽  
Vol 13 (3) ◽  
pp. 1248 ◽  
Author(s):  
Solange Cavalcanti de Melo ◽  
José Coelho de Araújo Filho ◽  
Renata Maria Caminha Mendes de Oliveira Carvalho
Keyword(s):  
Suspended Sediment ◽  
Suspended Sediment Concentration ◽  
Suspended Sediments ◽  
Sediment Concentration ◽  
Gradual Reduction ◽  
Sao Francisco River ◽  
Sediment Loads ◽  
São Francisco River ◽  
Historical Series ◽  
Suspended Sediment Concentrations

RESUMOO conhecimento da análise quantitativa das concentrações de sedimentos em suspensão transportados pelo rio São Francisco bem como sua relação com as vazões é de muita importância, pois pode auxiliar na identificação dos efeitos da intervenção humana e ou ocasionados pelas condições naturais da região. As regiões a jusante dos barramentos no rio São Francisco apresentam como principal consequência a regularização das vazões e a diminuição das concentrações de sedimentos. O objetivo da pesquisa foi determinar as curvas-chave de sedimentos em suspensão (CCS) nas estações fluviométricas instaladas no Baixo São Francisco (BSF) após a barragem de Xingó. Para o estabelecimento dessas curvas foram utilizados dados de vazão e concentração de sedimentos em suspensão, obtidos do sistema Hidroweb no site da Agência Nacional da Água (ANA) no período de 1999 a 2018. Foram obtidas CCS para todo o trecho do BSF as quais apresentaram bons coeficientes de determinação. Na análise dos dados também foi possível perceber que nos últimos anos, desde 2013 houve redução gradativa das vazões disponibilizadas na barragem de Xingó. Consequentemente, houve também a redução gradativa das cargas de sedimentos em suspensão geradas nas estações de Piranhas, Traipu e Propriá, ou seja, os menores valores já registrados no BSF correspondendo as menores séries históricas tanto de vazão como de sedimentos em suspensão.  Keys curves of sediment discharges in suspension in the Lower São Francisco A B S T R A C TThe knowledge of the quantitative analysis of suspended sediment concentrations carried by the São Francisco River as well as its relation with the flows is of great importance, since it can help in the identification of the effects of human intervention and/or caused by the natural conditions of the region. In the downstream regions of the São Francisco riverbanks, the main consequence was the regularization of flow rates and the reduction of sediment concentrations. The objective of the research was to determine the key curves of suspended sediments (CCS) at the fluviometric stations installed in the lower São Francisco river after Xingó dam. For the evaluation, flow data and suspended sediment concentration were used. These data were obtained from the Hidroweb system on the website of the National Water Agency (ANA) from 1999 to 2018. CCS were plotted for all stretches and presented good coefficients of determination (R2). Based on the analysis of the data it was also possible to notice that in recent years, since 2013 there has been a gradual reduction of the flows available in the Xingó dam. Consequently, there was also a gradual reduction of suspended sediment loads generated at the Piranhas, Traipu and Propriá stations, that is, the lowest values already recorded in lower São Francisco, corresponding to the lower historical series of both discharge and suspended sediments.Keywords: dam, flow, sediments 

scholarly journals Estimation of Suspended Sediment Concentrations with Adcp in Danube River

2013 ◽  
Vol 61 (3) ◽  
pp. 232-240 ◽  
Author(s):  
Sándor Baranya ◽  
János Józsa
Keyword(s):  
Measurement Uncertainty ◽  
Suspended Sediment ◽  
Sediment Load ◽  
Empirical Relationship ◽  
Danube River ◽  
Suspended Sediment Transport ◽  
Model Parameters ◽  
Load Estimation ◽  
Suspended Sediment Concentrations ◽  
Sediment Load Estimation

Abstract An estimation procedure for suspended sediment concentrations based on the intensity of backscattered sound of acoustic Doppler current profilers (ADCP) is introduced in this paper. Based on detailed moving and fixed boat ADCP measurements with concurrent sediment sampling, we have successfully calibrated the estimation method for a reach of River Danube in Hungary, characterized by significant suspended sediment transport. The effect of measurement uncertainty and various data filtering on sediment load determination is also analyzed and quantified. Some of the physical model parameters describing the propagation of sound in water are estimated based on known empirical formulas, while other parameters are derived from measured. Regression analysis is used to obtain a relationship between the intensity of backscattered sound and sediment concentrations. The empirical relationship has been then used to estimate the suspended sediment concentrations from the ADCP data collected in fixed and moving boat measurement operation mode, along verticals and path-lines, respectively. We show that while some measurement uncertainty is inherent to the acoustic Doppler principle, it is further enhanced by the complexity of the near-bottom sediment-laden flow. This uncertainty has then a significant effect on the local sediment load estimation. In turn, reasonable smoothing of raw velocity and backscatter intensity data shows insignificant impact on cross-sectional sediment load estimation.

scholarly journals Scale-breaks of suspended sediment rating in large rivers in Germany induced by organic matter

2020 ◽  
Author(s):  
Thomas O. Hoffmann ◽  
Yannik Baulig ◽  
Helmut Fischer ◽  
Jan Blöthe
Keyword(s):  
Organic Matter ◽  
Suspended Sediment ◽  
Suspended Matter ◽  
Monitoring Network ◽  
Suspended Sediments ◽  
Sediment Supply ◽  
Organic Fraction ◽  
Sediment Rating Curve ◽  
Suspended Sediment Concentrations ◽  
Rating Model

Abstract. Understanding the dynamics of suspended sediment and associated nutrients is of major relevance for sustainable sediment management aiming to achieve healthy river systems. Sediment rating curves are frequently used to analyze the dynamics of suspended sediments and their potential sources and sinks. Here we are using more than 750 000 measurements of the suspended sediment concentrations (SSC) and discharge at 62 gauging stations along 19 waterways in Germany based on the suspended sediment monitoring network of the German water and shipping authority, which started in the 1960ties. Furthermore, we analyse more than 2000 measurements of the loss on ignition (LOI) of suspended matter at two stations along the rivers Moselle and Rhine to asses the mineral and organic fraction of the suspended matter. SSC and LOI are analysed in terms of the power law rating to identify discharge depended process regimes of suspended matter. Our results indicate that for most studied gauging stations, rating coefficients are not constant over the full discharge range, but there is a distinct break in the sediment rating curve, with specific SSC-Q domains above and below this break. The transition of the rating exponent is likely to be a result of a change of controlling factors of the suspended sediment from intrinsic organic matter formation at low flows to extrinsic sediment supply (including mineral and organic fractions) due to hillslope erosion at high flows. Based on these findings we developed a conceptual rating model separating the mineral and organic fraction of the suspended matter in the Germany waterways. This model allows evaluating the sources of the mineral and organic fraction of the suspended matter and gain new insights into the first order control of discharge dynamics of suspended sediments.

scholarly journals Prediction of suspended-sediment concentrations at selected sites in the Fountain Creek watershed, Colorado, 2008-09

2013 ◽  
Author(s):  
Robert W. Stogner, Sr. ◽  
Jonathan M. Nelson ◽  
Richard R. McDonald ◽  
Paul J. Kinzel ◽  
David P. Mau
Keyword(s):  
Suspended Sediment ◽  
Creek Watershed ◽  
Suspended Sediment Concentrations

scholarly journals Monitoring of seasonal variability and movement of suspended sediment concentrations along the Thiruvananthapuram coast, southern India, using the Landsat OLI sensor

Ocean Science ◽  
2018 ◽  
Vol 14 (5) ◽  
pp. 1085-1092 ◽  
Author(s):  
Bismay Ranjan Tripathy ◽  
Kaliraj Seenipandi ◽  
Haroon Sajjad ◽  
Pawan Kumar Joshi ◽  
Bhagwan Singh Chaudhary ◽  
...  
Keyword(s):  
Suspended Sediment ◽  
Coastal Water ◽  
Seasonal Variability ◽  
Spectral Reflectance ◽  
Suspended Sediments ◽  
Southern India ◽  
Landsat 8 ◽  
Post Monsoon ◽  
Suspended Sediment Concentrations ◽  
Multi Temporal

Abstract. Studies on suspended sediment concentrations at a seasonal scale play a vital role in understanding coastal hydrodynamic processes in an area. Assessment of spatio-temporal changes in suspended sediments in nearshore areas has gained complexity due to the utilization of conventional methods; this issue can be successfully solved nowadays using multi-temporal remotely sensed images with the help of advanced image processing techniques. The present study is an attempt to demonstrate the model algorithm used to extract suspended sediment concentrations using Landsat 8 OLI (Operational Land Imager) sensor images. The study was executed in a near-offshore area of the Thiruvananthapuram coast, southern India, and focused on the extraction of suspended sediment concentrations and their seasonal variability during pre-monsoon and post-monsoon periods. The OLI images were pre-processed to obtain the actual reflectance using the FLASSH module of the ENVI v5.5 software. The generic model developed herein is designed to compute the spectral reflectance variability between coastal water and suspended sediments and to differentiate the spatial accumulation of the suspended sediment concentrations from the coastal water at the pixel scale. Maximum (0.8 % in near-infrared bands) and minimum (0.6 % in blue bands) spectral reflectance indicates the occurrence of suspended sediments in the coastal water. The model-derived results revealed that the suspended sediment concentration gradually decreased with increasing depth and distance from the shoreline. Higher sediment concentrations accumulated at lower depths in coastal water due to wave and current action that seasonally circulated the sediments. This higher concentration of the suspended sediment load was estimated to be 0.92 mg L−1 at the shallow depths (<10 m) of the coastal waters and 0.30 mg L−1 at a depth of 30 m. Seasonal variability of suspended sediments was observed in a north–south direction during the pre-monsoon; the reverse was noted during the post-monsoon period. The spatial variability of suspended sediments was indirectly proportional to the depth and distance from the shoreline, and directly proportional to offshore wave and littoral current activity. This study proves that the developed model coupled with the provided computational algorithm can be used as an effective tool for the estimation of suspended sediment concentrations using multi-temporal OLI images; furthermore, the output may be helpful for coastal zone management and conservation planning and development.

scholarly journals Suspended sediment load prediction using long short-term memory neural network

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Nouar AlDahoul ◽  
Yusuf Essam ◽  
Pavitra Kumar ◽  
Ali Najah Ahmed ◽  
Mohsen Sherif ◽  
...  
Keyword(s):  
Neural Network ◽  
Suspended Sediment ◽  
Short Term Memory ◽  
Suspended Sediments ◽  
Gradient Boosting ◽  
Short Term ◽  
Discharge Data ◽  
Term Memory ◽  
Extreme Gradient Boosting ◽  
Long Short Term Memory

AbstractRivers carry suspended sediments along with their flow. These sediments deposit at different places depending on the discharge and course of the river. However, the deposition of these sediments impacts environmental health, agricultural activities, and portable water sources. Deposition of suspended sediments reduces the flow area, thus affecting the movement of aquatic lives and ultimately leading to the change of river course. Thus, the data of suspended sediments and their variation is crucial information for various authorities. Various authorities require the forecasted data of suspended sediments in the river to operate various hydraulic structures properly. Usually, the prediction of suspended sediment concentration (SSC) is challenging due to various factors, including site-related data, site-related modelling, lack of multiple observed factors used for prediction, and pattern complexity.Therefore, to address previous problems, this study proposes a Long Short Term Memory model to predict suspended sediments in Malaysia's Johor River utilizing only one observed factor, including discharge data. The data was collected for the period of 1988–1998. Four different models were tested, in this study, for the prediction of suspended sediments, which are: ElasticNet Linear Regression (L.R.), Multi-Layer Perceptron (MLP) neural network, Extreme Gradient Boosting, and Long Short-Term Memory. Predictions were analysed based on four different scenarios such as daily, weekly, 10-daily, and monthly. Performance evaluation stated that Long Short-Term Memory outperformed other models with the regression values of 92.01%, 96.56%, 96.71%, and 99.45% daily, weekly, 10-days, and monthly scenarios, respectively.

scholarly journals Urban Flood Prediction Using Deep Neural Network with Data Augmentation

Water ◽  
2020 ◽  
Vol 12 (3) ◽  
pp. 899 ◽  
Author(s):  
Hyun Il Kim ◽  
Kun Yeun Han
Keyword(s):  
Neural Network ◽  
Urban Areas ◽  
Input Data ◽  
Deep Neural Network ◽  
Data Augmentation ◽  
Numerical Models ◽  
Training Data ◽  
Data Driven ◽  
Statistical Characteristics ◽  
Flood Prediction

Data-driven models using an artificial neural network (ANN), deep learning (DL) and numerical models are applied in flood analysis of the urban watershed, which has a complex drainage system. In particular, data-driven models using neural networks can quickly present the results and be used for flood forecasting. However, not a lot of data with actual flood history and heavy rainfalls are available, it is difficult to conduct a preliminary analysis of flood in urban areas. In this study, a deep neural network (DNN) was used to predict the total accumulative overflow, and because of the insufficiency of observed rainfall data, 6 h of rainfall were surveyed nationwide in Korea. Statistical characteristics of each rainfall event were used as input data for the DNN. The target value of the DNN was the total accumulative overflow calculated from Storm Water Management Model (SWMM) simulations, and the methodology of data augmentation was applied to increase the input data. The SWMM is one-dimensional model for rainfall-runoff analysis. The data augmentation allowed enrichment of the training data for DNN. The data augmentation was applied ten times for each input combination, and the practicality of the data augmentation was determined by predicting the total accumulative overflow over the testing data and the observed rainfall. The prediction result of DNN was compared with the simulated result obtained using the SWMM model, and it was confirmed that the predictive performance was improved on applying data augmentation.

scholarly journals Urban Street Pollutants

2008 ◽  
Vol 35 (2) ◽  
pp. 3 ◽  
Author(s):  
Cristiano POLETO ◽  
GUSTAVO HENRIQUE MERTEN
Keyword(s):  
Suspended Sediment ◽  
Urban Areas ◽  
Anthropogenic Activities ◽  
Suspended Sediments ◽  
Metropolitan Region ◽  
Runoff Water ◽  
Urban Street ◽  
Street Sediments ◽  
High Enrichment ◽  
High Concentrations

The sediments carried by runoff water are an important part of this process because their presence in the bodies of water not only cause sedimentation problems but, mainly, they contaminate the water due to the presence of the pollutants found associated with sediments. The urban subwatershed under study is located in the metropolitan region of Porto Alegre city, Brazil. This paper aims to present the relation between pollutants on the street sediments and in suspended sediment sampled in the river. The collections of suspended sediment samples begun in the end of 2003 and finished in the end of 2006. Collections of urban dust samples (47 samples per km²) were taken in the main diffuse sources of the urban environment, represented by paved and non-paved streets, beyond the area with remaining vegetation, in some points of the bed river and in its margins. During these analyses, it was studied 29 samples of fluvial suspended sediments. The elements selected for this study are some of the most frequently found in high concentrations in urban areas (Zn, Pb and Cu). The results suggest it is occurring a high enrichment of the local sediment with these metals. The concentrations of these elements vary temporally during storms due the input of road runoff containing elevated concentrations of elements associated with vehicular traffic and other anthropogenic activities. In general, they have their most concentrations on the streets but they are carried to the channel during the storms.

scholarly journals Dynamically Constrained Interpolation of the Sparsely Observed Suspended Sediment Concentrations in Both Space and Time: A Case Study in the Bohai Sea

2018 ◽  
Vol 35 (5) ◽  
pp. 1151-1167 ◽  
Author(s):  
Xinyan Mao ◽  
Daosheng Wang ◽  
Jicai Zhang ◽  
Changwei Bian ◽  
Xianqing Lv
Keyword(s):  
Suspended Sediment ◽  
Bohai Sea ◽  
Transport Model ◽  
Suspended Sediment Transport ◽  
Model Parameters ◽  
Constrained Interpolation ◽  
The Bohai Sea ◽  
Space And Time ◽  
Twin Model ◽  
Suspended Sediment Concentrations

AbstractThe observed suspended sediment concentrations (SSCs) obtained from the water sampling are usually sparsely distributed in both space and time, which are traditionally applied just to calibrate other types of observations. In this study a dynamically constrained interpolation methodology (DCIM) is developed to interpolate these sparsely observed SSCs in the Bohai Sea. In this method the suspended sediment transport model is taken as dynamical constraints to interpolate the observations. Meanwhile, the interpolated results are optimized iteratively by adjusting the key model parameters using the adjoint method.The DCIM is first verified using the synthetic observations produced by twin model runs. The modeling results reveal that this method is effective at interpolating the sparsely observed artificial SSCs, even when the observations are heavily contaminated by data noise. Then, the sparsely observed practical SSCs obtained from a large area survey in the Bohai Sea are interpolated using the DCIM. The interpolated results are verified by randomly selected independent observations. The discrepancies between the interpolated SSCs and the observations are significantly decreased. When all the observations are interpolated, the final interpolated SSCs captured a majority (96.88%) of observations with a factor of 2 and the correlation coefficient between the observed and interpolated SSCs is 0.98. Besides, the interpolated results have presented the reasonable dynamical variations of SSCs in the space and time domains. The modeling results indicate that the DCIM is an effective tool for interpolating the sparsely observed SSCs in both space and time.

scholarly journals A new approach for estimation suspended sediment yield

1999 ◽  
Vol 3 (2) ◽  
pp. 285-294 ◽  
Author(s):  
R. Lidén
Keyword(s):  
Suspended Sediment ◽  
Suspended Sediment Concentration ◽  
Sediment Concentration ◽  
Model Parameters ◽  
Rating Curve ◽  
Sediment Rating Curve ◽  
Daily Series ◽  
Suspended Sediment Concentrations ◽  
Model Equations

Abstract. A semi-distributed conceptual model, HBV-SED, for estimation of total suspended sediment concentration and yield at the outlet of a catchment was developed and tested through a case study. The base of the suspended sediment model is a dynamic hydrological model, which produces daily series of areal runoff and rainfall for each sub-basin as input to the sediment routine. A lumped measure of available sediment is accumulated continuously based on a linear relationship between log-transformed values of rainfall and erosion, while discharge of suspended sediment at the sub-basin outlet is dependent on runoff and amount of stored available sediment. Four model parameter are empirically determined through calibration against observed records of suspended sediment concentration. The model was applied to a 200 km2 catchment with high altitude differences in the tropical parts of Bolivia, where recorded suspended sediment concentrations were available during a two-year period. 10,000 parameter sets were generated through a Monte Carlo procedure to evaluate the parameter sensitivity and interdependence. The predictability of the model was assessed through dividing the data record into a calibration and an independent period for which the model was validated and compared to the sediment rating curve technique. The results showed that the slope coefficients of the log-transformed model equations for accumulation and release were much stronger than the intercept coefficients. Despite and existing interdependence between the model parameters, the HBV-SED model gave clearly better results than the sediment rating curve technique for the validation period, indication that the supply-based approached has a promising future as a tool for basic engineering applications.

Sign in / Sign up

Export Citation Format

Share Document