Premium
Gene Expression Programing for Estimating Suspended Sediment Yield in Middle Euphrates Basin, Turkey
Author(s) -
Guven Aytac,
Talu Necip Ersin
Publication year - 2010
Publication title -
clean – soil, air, water
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.444
H-Index - 66
eISSN - 1863-0669
pISSN - 1863-0650
DOI - 10.1002/clen.201000003
Subject(s) - gene expression programming , sediment , regression analysis , variable (mathematics) , regression , yield (engineering) , mathematics , statistics , structural basin , rating curve , drainage basin , linear regression , variables , hydrology (agriculture) , expression (computer science) , environmental science , geology , geography , computer science , geotechnical engineering , mathematical analysis , cartography , geomorphology , physics , machine learning , programming language , thermodynamics
Gene expression programing (GEP) is used to estimate the suspended sediment yield (SSY) in Euphrates River. SSY is considered to be a function of (i) discharge and (ii) time‐lagged discharge and SSY. The proposed models were trained to extrapolate natural stream data collected from five stations in Middle Euphrates Basin. A detailed sensitivity analysis is done to select the time‐lagged discharge and sediment yield variables. GEP implicitly evaluates the contribution of each independent variable on the fitness of candidate solution and eliminates the variable having no contribution. In this study, all input variables are observed to be included in the proposed GEP models, which prove the significance of each variable. Also, standard and modified sediment rating curves and regression‐based formulae are developed for the five stations. In verification, the estimations of GEP formulae agree well with the measured ones. The GEP models are evaluated by the results of the rating curves and regression formulae. In general, the GEP formulae give better results compared to the rating curves and regression‐based formulae.