HR: 14:30h
AN: H43A-03    [Abstracts]
TI: Suspended Sediment Load Prediction Using Artificial Neural Networks Approach
AU: * Melesse, A
EM: melessea@fiu.edu
AF: Department of Environmrntal Studies Florida International University, 11200 sw 8th st, Miami, FL 33199, United States
AU: Ahmad, S
EM: sajad.ahmad@unlv.edu
AF: Department of Civil and Environmental Engineering University of Nevada, Las Vegas , 4505 Maryland Parkway , Las Vegas, NV 89154-4015, United States
AU: McClain, M
EM: mcclainm@fiu.edu
AF: Department of Environmrntal Studies Florida International University, 11200 sw 8th st, Miami, FL 33199, United States
AU: Wang, X
EM: xixi.wang@und.nodak.edu
AF: Environmental and Energy Research Center, University of North Dakota, 15 North 19rd Stree, Grand Forks, ND 58202, United States
AU: Lim, H
EM: lim@und.edu
AF: Department of civil Engineering, University of North Dakota, 4149 University Ave, Grand Forks, ND 58202, United States
AU: Nangia, V
EM: nangia@umac.org
AF: UMAC, University of North Dakota, 4149 University Ave, Grand Forks, ND 58202, United States
AB: A multilayer perceptron (MLP) ANN with an error back propagation algorithm using historical daily and weekly hydroclimatological data (precipitation P(t), current discharge Q(t), antecedent discharge Q(t-1), and antecedent sediment load SL(t-1) ) is used to predict the suspended sediment load SL(t) at the selected monitoring stations. Performance of ANN was evaluated using different combinations of datasets (Input 1 = P(t), Q(t), Q(t-1), SL(t-1) , Input 2 = I-1 less P(t) and Input 3 = I-2 less Q(t-1), length of record for training (3 and 2 years) and temporal (daily and weekly) simulations. Comparison of the ANN model output with multiple linear regressions (MLR) was made. Daily simulations using Input 1 and three years of training and two years of testing (3*2) performed better (R2 and E of 0.85 and 0.72, respectively ) than the simulation with two years of training and three years of testing (2*3) (R2 and E of 0.64 and 0.46, respectively ). ANN predicted daily values using Input 1 and 3*2 architecture for Missouri (R2 = 0.97) and Mississippi (R2 = 0.96) were better than Rio Grande (R2 = 0.65). Daily predictions were better compared to weekly predictions for all three rivers. ANN predictions for Missouri and Mississippi were superior and Rio Grande was inferior compared to predictions with MLR. The modeling approach presented in this paper can be an efficient alternative to costly monitoring operations for sediment load monitoring programs where hydrological data is readily available. Key terms: ANN, sediment, sediment prediction, rivers, Mississippi, Missouri, Rio Grande
DE: 1815 Erosion
DE: 1847 Modeling
DE: 1861 Sedimentation (4863)
SC: Hydrology [H]
MN: 2007 Joint Assembly