Accurate Modeling of Suspended Sediment Using a Hybrid Intelligent Approach Based on Sediment Rating Curve and CANFIS Network (Case Study: Glinak Station) | ||
| مهندسی و مدیریت آبخیز | ||
| Articles in Press, Accepted Manuscript, Available Online from 08 July 2026 | ||
| Document Type: Research Paper | ||
| DOI: 10.22092/ijwmse.2026.370028.2123 | ||
| Author | ||
| Golaleh Ghaffari* | ||
| Assistant Professor, Department of Soil Conservation and Watershed Management, Research and Education Center for Agriculture and Natural Resources, Kermanshah Province, Iran | ||
| Abstract | ||
| Accurate modeling of suspended sediment is a key challenge in water resources management, requiring methods capable of capturing the complex and nonlinear behavior of sediment transport processes. This study investigates a hybrid modeling approach that integrates sediment rating curves (SRC) and artificial neural networks (ANNs) to predict suspended sediment load at the Glink hydrometric station in the Taleghan watershed. Historical streamflow and suspended sediment data from 1971 to 2023 were preprocessed and analyzed using three modeling frameworks: (i) empirical models (six types of SRCs), (ii) data-driven models (MLP, GFF, RBF, SVM, SOFM, and CANFIS), and (iii) two hybrid structures. Among the empirical models, the "class-mean method" outperformed others with the highest coefficient of determination (R² = 0.75) and the lowest relative mean error (RME), making it the most reliable rating curve approach. The results unequivocally demonstrate that the synergistic integration of empirical hydrological reasoning and artificial intelligence provides a resilient and efficient modeling framework for watersheds characterized by pronounced skewness and a high prevalence of anomalous observations. In this context, the CANFIS architecture, leveraging fuzzy logic to resolve complex nonlinear interactions, substantially improves the representation of hysteretic sediment dynamics and offers clear advantages over conventional computational structures, including multilayer perceptron (MLP), general feed-forward (GFF), radial basis function (RBF), support vector machine (SVM), and self-organizing feature map (SOFM) networks. Given the strategic significance of flood events in water resources engineering, the rigorous interrogation of outlier behavior is not optional but imperative. The proposed hybrid framework effectively captures and interprets these statistical anomalies with acceptable precision, thereby reinforcing its applicability in suspended sediment transport management under extreme hydrological conditions. To advance methodological robustness and scientific depth in future investigations, the establishment of comprehensive, high-resolution databases—particularly those documenting peak flood discharges—is strongly recommended. | ||
| Keywords | ||
| Suspended Sediment; Sediment Rating Curve; Artificial Neural Network; Hybrid Modeling; CANFIS; Taleghan Watershed | ||
| References | ||
|
| ||
|
Statistics Article View: 9 |
||
Journal management system. designed by sinaweb