Performance of Six Methods in Determining Solute-Transport Parameters from Breakthrough Curves

Abstract

ABSTRACT The solute-transport parameters are usually determined by fitting solute-transport models to measured solute breakthrough curves (BTCs). But due to different initial and boundary conditions and approximations, the solute-transport models perform with different accuracy levels, making it difficult to select the most appropriate one for a particular type of breakthrough experiment. This study evaluated the performances of four commonly used solute-transport models and two methods by analyzing the BTCs of 81 pulse-type breakthrough experiments with an inert solute (CaCl2) in three different soil-types and identified their strengths and weaknesses. The employed models were: the equilibrium (EQ) and non-equilibrium (NEQ) models of the CXTFIT codes, and analytical solutions of Lapidus and Amundson (LA) and Lindstrom and coauthors (LI), while the two methods were: the method of moments (MOM), and transfer-function (TF) method, the latter still remains less studied. The MOM accurately determined mean velocity (V) and travel time () of the solute but poorly determined its mass-dispersion number (N), dispersion coefficient (D), and dispersivity (); note that, for inert solutes, V is identical to the mean pore-water velocity. The NEQ model always provided inconsistent solute-transport parameters, while the LA and LI models sometimes provided inconsistent parameters; all three models suffered from the problems of convergence during the fitting process. The EQ model and the TF method resulted in the consistent solute-transport parameters, with the latter method performing better. The results of this study will guide to select appropriate model/method to determine reliable solute-transport parameters from pulse-type solute-transport experiments.