The input file is "tracer.in", the superfile is "hbgc123d.tra".
This sample demonstrates the use of HBGC123D for a simple transport problem without chemical or microbiological reactions. The model domain represents a laboratory column experiment. The measurements of the cylindrical column are length = 14.5 cm and radius = 4.2 cm. This is represented by a one-dimensional mesh with 58 elements and 59 nodes. The porosity is 0.428, the bulk density of the soil is r = 1.428 kg/dm3.
Initially, the column contained no tracer. The inflow of tracer (bromide in the experiment) began at the start of the experiment and lasted for 1826.5 h. The inflow concentration is 3.6 mmol/l. Afterwards, a washout phase began and no tracer flowed into the column anymore. Two flow interrupts were performed in the course of the experiment. The first one started at t = 1267.45 h and lasted until t = 1508.45 h. The second one started at t = 3155.5 h and lasted until t = 3342.5 h. During these flow interrupts, no water or tracer flowed into the column. During the remainder of the experiment the inflow rate into the column was q = 5 ml/h. This yields a Darcy velocity of v = 9.022*10-3 dm/h. The units used in the HBGC123D simulation are dm, h, mol. A time step size of Dt = 0.05 h was used in this simulation.
In order to simulate the various flow phases (flow and flow interrupts), KVI was set to -2 in Data Set 2. 10 data points were specified in Data Set 16 for both velocity and water content. The water content is constant during the entire experiment, so that these values are the same for all data points.
Since there is only one component, Data Sets 25 through 28 are not needed for this simulation and only a "0" as the number of reactions has to be specified in Data Set 29.
Figure 1 shows the breakthrough curve (concentration in the effluent from the column in the experiment and concentration at node 59 in the simulation) for Bromide. The agreement between the simulated and the measured data is very good. Since there are no reactions involved, only the longitudinal dispersivity (PROP(1,I) in Data Set 6 needed to be adjusted.
Figure 1: Breakthrough curve for Bromide, comparison of simulated and measured data.
Acknowledgement
The column data were provided by Philip Jardine and Melanie Mayes of the Environmental Sciences Divison of Oak Ridge National Laboratory. The experimental results will be published in Mayes et al. (1999).
Reference
Mayes, M. A., Jardine, P. M., Larsen, I. L., Brooks, S. C., and Fendorf, S. E. (1999): Multispecies contaminant transport in undisturbed columns of weathered, fractured saprolite". Submitted to Journal of Contaminant Hydrology, July 1999 (in review).