Zatlakovič M., Hodasová K. & Krčmář D., 2021: Solute transport by groundwater – a comparison of computation methods sensitivity to physical-chemical parameters and source geometry. Acta Geologica Slovaca, 13, 1, 95–106.

Solute transport by groundwater – a comparison of computation methods sensitivity to physical-chemical parameters and source geometry

Martin Zatlakovič, Kamila Hodasová & Dávid Krčmář

Department of Engineering Geology, Hydrogeology and Applied Geophysics, Faculty of Natural Sciences, Comenius University in Bratislava, Ilkovičova 6, 842 15 Bratislava, Slovakia; martin.zatlakovic@uniba.sk

Abstract

The presented comparative study evaluates the sensitivity of selected computational procedures to chosen parameters of solute transport and source geometry in homogeneous conditions. The computation procedures include three numerical schemes with FDI (finite difference implicit), MOC (method of characteristics), and a TVD (total variation diminishing) advection solver, while the other terms of the governing equation are computed by the FDI method. Furthermore, two analytical solutions were used: the exact solution of Newille (2005) and the approximate solution of Domenico (1987). Finally, a simplified, so-called Step-method, which is still being used in the practice, was used. The sensitivities of the individual procedures to selected physical-chemical parameters and to the geometric characteristics of the solute source were evaluated and compared. The obtained results show considerable differences between the particular procedures used. In most test cases, the numerical procedures with a TVD and MOC advection solver produced higher concentrations and were more sensitive to the transport parameters in comparison with the other methods used. Applied variations of individual parameter values caused a change of the calculated solute concentrations up to 7.1 %, whereas the substitution of calculation methods caused up to 7.7 % of C0 (the concentration in solute source). The simultaneous changes of all the examined parameter values with a synergistic effect caused the maximal change of calculated concentration up to 15.3 % of C0. A significant effect of the solute source geometry on the match between the results obtained by the selected numerical and analytical methods was also found (difference up to 57.9 % of C0). The achieved results indicate a need to solve the solute transport in homogeneous conditions using both types of calculation methods, the numerical and analytical, in order to cover possible variability of resulting concentrations. The simplified Step-method predominantly yields the lowest concentrations. At the same time, it is the least sensitive procedure to the selected parameters and can be used only for a reference point situated on the plume centerline. From a cautious point of view, it is the least suitable method for solute transport modelling among the compared methods.

Key words: solute transport, analytical model, numerical model, transport parameters, method sensitivity

Manuscript received: 2021-02-15

Revised version accepted: 2021-05-25