Steady-State MT3DMS

[Phillip Goodling]

Hello All, 

I am modeling non-reactive contaminants leaving a swamp and entering a river in a steady-state Modflow 2005 simulation using ModelMuse. I would like to show these contaminants diluted by dispersion, advection, and incoming precipitation along the flow path. 

The packages I am using are:

- Constant Head Boundary (for swamp and river steady-state model)

- Recharge (for precipitation infiltration)

- MT3DMS: the Basic Transport, Advection, Dispersion, Sink and Source Mixing, and Generalized Conjugate Gradient Solver packages

My I have found the dispersion coefficient of the contaminant in the literature and am adjusting the longitudinal dispersivity to fit my observed field data. My lake has a constant concentration and precipitation has a concentration of zero.

Here are my issues:

1) I seem to be losing my contaminant, as I always have a large in/out discrepancy in the listing file. The concentrations in my plume are also far lower than I expect no matter how I change the dispersivity. 

2) The cells not in the plume (would expect a concentration of 0) read a negative concentration of contaminant. Which is patently impossible. 

Does anyone know if I am conceptualizing this problem correctly or where my contaminant is disappearing to? Ay insight would be most welcome. I have been reading page 74 of the MT3DMS manual, but cannot tell if decreases in mass balance due to storage applies in a steady-state model, where the head is not changing. 

Thanks!

-Phillip

[Nadège Vanden Berghe]

Hi Philip

You are likely facing numerical issues. Have you tested different transport solver and time settings ?