One-dimensional numerical modelling of solute transport in streams: The role of longitudinal dispersion coefficient

- 10 empirical formulas of longitudinal dispersion coefficient are investigated.
- Solute transport under steady and unsteady flow conditions is simulated numerically.
- A 1-D numerical model is used.
- Iwasa and Aya (1991) formula is the best predictor.

One-dimensional (1-D) numerical models of solute transport in streams rely on the advection-dispersion equation, in which the longitudinal dispersion coefficient is an unknown parameter to be calibrated. In this work we investigate the extent to which existing empirical formulations of longitudinal dispersion coefficient can be used in 1-D numerical modelling tools of solute transport under steady and unsteady flow conditions. The 1-D numerical model used here is the open source Mascaret tool. Its relevance is illustrated by simulating theoretical cases with known analytical solutions. Ten empirical formulas of longitudinal dispersion coefficient are then tested by simulating eight laboratory experimental cases under steady flow condition and the solute transport in the Middle Loire River (350 km long) under highly variable flow condition (from July 1st 1999 to December 31st 1999). Comparisons between computed and measured breakthrough curves show that Elder (1959), Fischer (1975) and Iwasa and Aya (1991) formulas rank as the best predictors for the experimental cases. For the field case, Seo and Cheong's (1998) formula yields the best model-data agreement, followed by Iwasa and Aya's (1991) formula. The latter formula is, therefore, recommended for the entire range of conditions studied here.