MT3D-USGS works with FD solver but fails with TVD solver

[Ryan Conway]

Hello,

I have a 200 year MT3D-USGS model which runs and converges with a good mass balance (creeps up to around 0.5% around 100 years then creeps back down to 0.1%) using the finite difference method with a uniform transport timestep of 3 days. My flow model is transient, and cells are drying out and re-wetting. 

However, I tried to switch to TVD (I understand that it is the "best" ADE solver, especially when transport is advection dominated, as mine should be) and the model fails to converge right away.

I tried decreasing PERCEL in .adv as well (to as little as 0.01) and it still fails. Here is a screen shot of the command window:

Next I tried reducing the the timestep in the FD solver. Now, the model runs and converges and results look pretty much the same, but the .out file is reporting huge masses for my sorbing component (liner isotherm). Here is a screenshot of the .out file:

Any ideas whats going on here? Do drying/rewetting cells cause weird issues in MT3D-USGS?

Thanks,

 

[Ashutosh]

This issue has been discussed in this forum quite a lot. Look at the previously discussed mt3d and dry cell problems in the same forum. You will find contact of people who have worked on these issues and might have a go around for it.

//Hpc

Sent from my iPhone

On 3 Jun 2019, at 19:36, Ryan Conway <rcon...@gmail.com> wrote:

Hello,

I have a 200 year MT3D-USGS model which runs and converges with a good mass balance (creeps up to around 0.5% around 100 years then creeps back down to 0.1%) using the finite difference method with a uniform transport timestep of 3 days. My flow model is transient, and cells are drying out and re-wetting. 

However, I tried to switch to TVD (I understand that it is the "best" ADE solver, especially when transport is advection dominated, as mine should be) and the model fails to converge right away.

I tried decreasing PERCEL in .adv as well (to as little as 0.01) and it still fails. Here is a screen shot of the command window:

<tvd_error.PNG>

Next I tried reducing the the timestep in the FD solver. Now, the model runs and converges and results look pretty much the same, but the .out file is reporting huge masses for my sorbing component (liner isotherm). Here is a screenshot of the .out file:

<mt3d_mass_errors.PNG>

Any ideas whats going on here? Do drying/rewetting cells cause weird issues in MT3D-USGS?

Thanks,

 

--
You received this message because you are subscribed to the Google Groups "MODFLOW Users Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to modflow+u...@googlegroups.com.
To post to this group, send email to mod...@googlegroups.com.
Visit this group at https://groups.google.com/group/modflow.
To view this discussion on the web visit https://groups.google.com/d/msgid/modflow/6d7a9e29-a5d5-4d66-b28c-68ffbb89216d%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

<tvd_error.PNG>

<mt3d_mass_errors.PNG>