Experimental Investigation and Evaluation Results of Numerical Simulation and Analytical Solution of classical ADE for conservative solutes

Subject Areas : Farm water management with the aim of improving irrigation management indicators

alireza  emadi ¹ (Department of Water Engineering, Sari Agricultural Sciences and Natural Resources University,sari,iran)
yaghoub  azhdan ² (Department of Water Engineering,Sari Agricultural Sciences and Natural Resources University,Sari, Iran.)
jafar  chabokpour ³ (Faculty of Engineering, University of Maragheh, Maragheh, Iran)
Rasoul  daneshfaraz ⁴ (Faculty of Engineering, University of Maragheh, Maragheh, Iran)

Received: 2018-05-22 Accepted : 2019-01-14 Published : 2018-12-22

Keywords: Longitudinal Dispersion Coefficient, Solute Transport, Breakthrough Curve, OTIS,

Abstract :

The longitudinal dispersion coefficient from the perspective of public safety and human health, is one of the important parameters for predicting and characterizing the transport of pollution in rivers and shallow waters. This paper aimed at investigating the efficiency of the classical advection- dispersion Equation for predicting transport of pollution in a gravel bed rivers. Some tracer experiments were done in a flume with longitudinal slopes of (S= 0.001, 0.007) and five discharges include of (7.5, 11.5, 15.5, 20.5, 25.5) lit/sec. Simulated (OTIS) and analytical breakthrough curves were compared with observed data. The results of this study indicated that advection- dispersion equation in prediction of tailing edge of breakthrough curve has less accuracy. So it not be used in rivers with storage area but it is very helpful to estimate initial value of longitudinal dispersion coefficient. The calculated dispersion coefficient increased with augmentation distance from the injection site. The domain of estimated dispersion coefficient for the tracer test of this research is between (0.0073- 0.18) m2/sec. the results of calculating travel time parameters from experimental BT curves showed an exponential relationship between them and the distance from the injection site. Finally, using a geometric, hydraulic, and tracer test data and applying a Buckinghamۥ π theorem a new equation was developed to predict the longitudinal dispersion coefficient. The relative error and normalized root mean square error for the proposed equation were calculated respectively 24% and 1.37.