Toward multi-day-ahead forecasting of suspended sediment concentration using ensemble models

Abstract

This study explores two ideas to made an improvement on the artificial neural network (ANN)-based models for suspended sediment forecasting in several time steps ahead. In this regard, both observed and forecasted time series are incorporated as input variables of the models when applied for more than one lead time. Secondly, least-square ensemble models employing multiple wavelet-ANN models are developed to increase the performance of the single model. For this purpose, different wavelet families are linked with the ANN model and performance of each model is evaluated using error measures. The Skagit River near Mount Vernon in Washington county is selected as the case study. The daily flow discharge and suspended sediment concentration (SSC) in the current day are considered as input variables to predict suspended sediment concentration in the next day. For more lead times, the input structure is updated by adding the forecast of SSC in the previous time step. Results of this study demonstrate that incorporating both observed and predicted variables in the input structure improves performance of conventional models in which those only employ observed time series as input variables. Moreover, ensemble model developed for each lead time outperforms the best single wavelet-ANN model which indicates superiority of the ensemble model over the other one. Findings of this study reveal that acceptable forecasts of daily suspended sediment concentration up to 3 days in advance can be achieved using the proposed methodology.

The assessment of water quality in river Molo water basin, Kenya

“…Nonetheless, it can predict the origin of recharge water and the source of water quality parameters. Acceptable forecasts of daily suspended sediment concentration of up to 3 days in advance can be predicted accurately using the proposed single-wavelet ANN model which indicates superiority of the ensemble model (Alizadeh et al 2017).…”

“…Studies such as the artificial neural network (ANN) model for suspended sediment forecasting in several time steps, soft computing methods such as adaptive neuro-fuzzy inference system (ANFIS) and AquaChem computational models have become necessary in investigating water quality in addition to the traditional experimental methods (Wang et al 2014;Wu and Chau 2006). The method employed in this study, however, does not measure suspended sediment load (SSL) which is indispensable for planning and management of water resource structures (Alizadeh et al 2017;Olyaie et al 2015). Nonetheless, it can predict the origin of recharge water and the source of water quality parameters.…”

Towards the Development of an Affordable and Practical Light Attenuation Turbidity Sensor for Remote Near Real-Time Aquatic Monitoring

“…Some proposals examine low-cost monitoring of turbidity in drinking water [15][16][17][18]. Some work has been undertaken on low-cost sensor designs [19], distributed real-time turbidity monitoring [20,21], and using machine learning techniques for predicting sediment load in waterways [22][23][24][25][26][27]. However, few designs specifically target the application of cost-effective remote environmental monitoring, nor are commercial turbidity sensors practical for remote in situ deployment over an extended period.…”

Groundwater Quality Patterns and Spatiotemporal Change in Depletion in the Regions of the Arabian Shield and Arabian Shelf

“…Potential and development of the Saq aquifer have been studied by different researchers [7][8][9][10][11][12][13][14][15]. In addition, several studies focused on water quality [16][17][18][19][20][21]. This area needs more research to sustain the Saq aquifer as the main water supply.…”