Model identification and accuracy for estimation of suspended sediment load

Abstract

In the present study, three widely used modeling approaches: (1) sediment rating curve (SRC) and optimized OSRC, (2) machine learning models (ML) (random forest (RF) and Dagging-RF (DA-RF)) and (3) the semi-physically based soil and water assessment tool (SWAT) are applied to predict suspended sediment load (Q_s) at the Talar watershed in Iran. Various graphical and quantitative methods were used to evaluate the goodness of fit. Results indicated that the RF model had the best prediction power in the training phase, while the dagging-RF hybrid algorithm outperformed all other models in the validation phase. The OSRC, RF and DA-RF had ‘very good’ performances based on the NSE in the validation phase, SRC showed ‘good’ performance, while the predicted values using SWAT were ‘satisfactory’. Our results suggest that the OSRC and ML models are more suitable for prediction of Q_s in study catchments with poor data availability.

Graphical Abstract

Keywords:

• Sediment rating curve
• machine learning
• SWAT
• OSRC
• Talar watershed
• Iran

Acknowledgements

The authors thank the I.R of Iran Meteorological Organization and the Mazandaran Regional Water Authority for providing datasets. The authors also acknowledge the support of Mrs. Yasaman Shokouhifar for her guidance and support during the implementation of SWAT, Dr. Maziar Mohammadi for sharing some of the datasets, Dr. Mohamad Ehteram for guidance on the use of the metaheuristic algorithm for the sediment rating curve optimization, and special thanks to Professor Karim C. Abbaspour for his review and comments which significantly enhanced the quality of the current paper.

Author’s contributions

Khabat Khosravi: Conceptualization, data curation, formal analysis, methodology, software, writing - original draft, review and editing. Ali Golkarian: Conceptualization, methodology, supervision, review and editing. Patricia M. Saco: Conceptualization, methodology, review and editing. Martijn J. Booij: Methodology, review and editing. Assefa Melesse: Conceptualization, supervision, review and editing

Data availability statement

Data shared with DOI of 10.4121/16685191 at https://figshare.com/s/e20fa8bbba586a173d1a

Disclosure statement

No potential conflict of interest was reported by the authors.

Declarations

Not applicable

Ethics approval

Not applicable.

Consent to participate

Not applicable.

Consent for publication

Not applicable.

Additional information

Funding

Khabat Khosravi was partially supported by a grant from the Ferdowsi University of Mashhad (No. FUM-1399102452809).