A comprehensive uncertainty analysis of model-estimated longitudinal and lateral dispersion coefficients in open channels

JOURNAL OF HYDROLOGY

卷:603

Article

Najafzadeh, Mohammad¹  Noori, Roohollah^2,6  Afroozi, Diako¹  Ghiasi, Behzad³  Hosseini-Moghari, Seyed-Mohammad⁴  Mirchi, Ali⁵  Haghighi, Ali Torabi²  Klove, Bjorn² 

[1] Grad Univ Adv Technol, Fac Civil & Surveying Engn, Dept Water Engn, Kerman 76315116, Iran

[2] Univ Oulu, Fac Technol, Water Energy & Environm Engn Res Unit, Oulu 90014, Finland

[3] Univ Tehran, Coll Engn, Sch Environm, Tehran 1417853111, Iran

[4] Chinese Acad Sci, Inst Geog Sci & Nat Resources Res, Key Lab Water Cycle & Related Land Surface Proc, Beijing 100101, Peoples R China

[5] Oklahoma State Univ, Dept Biosyst & Agr Engn, 111 Agr Hall, Stillwater, OK 74078 USA

[6] Iran Univ Sci & Technol, Sch Civil Engn, Tehran 1684613114, Iran

关键词: Open channel;    Machine learning models;    Rates of mixing;    Uncertainty;   

DOI  :  10.1016/j.jhydrol.2021.126850

来源: Elsevier

【 摘 要 】

The complexity of pollutant-mixing mechanism in open channels generates large uncertainty in estimation of longitudinal and lateral dispersion coefficients (K-x and K-y). Therefore, K-x and K-y estimation in rivers should be accompanied by an uncertainty analysis, a subject mainly ignored in previous studies. We introduce a method based on thorough analysis of different calibration datasets, resampled from a global database of tracer studies, to determine the uncertainty associated with five applicable intelligent models for estimation of K-x and K-y (model tree, evolutionary polynomial regression (EPR), gene expression programming, multivariate adaptive regression splines (MARS), and support vector machine (SVM)). Our findings suggest that SVM gives least uncertainty in both K-x and K-y estimation, while EPR and MARS generate most uncertainty in K-x and K-y estimation, respectively. By considering significant uncertainty in the model estimations, we suggest that the methodology we introduce here for uncertainty determination of the models be incorporated in empirical studies on estimation of K-x and K-y in rivers.

【 授权许可】

Free