Jump to content
GMS, SMS, and WMS User Forum
Matthew

Flooded Cells Beneath Dry Cells

Recommended Posts

:blink:

I have a MODFLOW steady state simulation where some of the top layer cells go dry in areas where this may truly be what is happening. However, the cells immediately below the dry cells are calculated as flooded, which is obviously inconsistent and nonsensical. Has anybody experienced this, and how did they solve it?

I have tried various solvers- the only ones that work are PCG and GMG. They seem to converge nicely despite the obvious problem. I have tried adjusting the starting heads to the solution, with heads set at the bottom of dry layers.

Any help would be appreciated likely beyond any appreciation you have ever received or imagined thus far in your life.

Share this post


Link to post
Share on other sites
:blink:

I have a MODFLOW steady state simulation where some of the top layer cells go dry in areas where this may truly be what is happening. However, the cells immediately below the dry cells are calculated as flooded, which is obviously inconsistent and nonsensical. Has anybody experienced this, and how did they solve it?

I have tried various solvers- the only ones that work are PCG and GMG. They seem to converge nicely despite the obvious problem. I have tried adjusting the starting heads to the solution, with heads set at the bottom of dry layers.

Any help would be appreciated likely beyond any appreciation you have ever received or imagined thus far in your life.

Matthew,

There is probably nothing wrong with the solution. The apparent flooding is an artefact of how "flooded" cells are displayed in GMS. If the head in the uppermost active cell is above the top of the cell, it will be marked as flooded. If this is the top layer of the model, then you have real flooding. In the case of lower cells, however, the overlying cell has probably run dry and been inactivated, but heads have then risen above the top of the underlying active cell. You may need to allow re-wetting so that the cell above can become active again, but of course that may lead to instability.

If you want to show which areas are flooded/artesian, you would be better of using DATA, 3D Data>2D Data and saving "Highest active value in ij column" In the 2D DATA CALCULATOR menu, subtract the ground surface from this head, which will give you the head above ground surface. Then contour it showing only positive values.

Share this post


Link to post
Share on other sites

Thanks for the reply Rob. My problem was that for a long time I thought I was allowing re-wetting properly- I seemed to have had all the right "allow re-wetting" boxes checked in the LPF dialog. So I couldn't understand why the cells above flooded cells were not re-wetting. It turned out that just checking the rewetting box and choosing the wetting equation from the drop-down box doesn't get the job done- apparently the wet/dry flag was still set at a default value (0) that doesn't allow re-wetting.

There was nothing useful in the GMS help file about the wet/dry flag, but I learned by web-searching that I needed to set it to a negative value to allow rewetting from beneath only, and that the value should be selected to be a non-zero number (its absolute value representing the threshold flooding thickness- with the sign chosen for the source cells for re-wetting- "cell beneath" is negative, "all adjacent is positive"- that triggers re-wetting, balanced for solution stability). I asked a lot of people for help on this, but no one pointed me in this direction until I finally stumbled on it- It may have been too obvious to most MODFLOW users! But I am only an occasional episodic GMS/MODFLOW user, so I did not notice or understand it.

I am still running into some errors and stability problems in the thinly-saturated cells, but at least I am finally making some sense out of the rewetting behavior. Unfortunately, it took me a very long time of hitting my head against the wall to discover the full meaning of the wet/dry flag and threshold. Maybe this thread will help others avoid the same frustration. I also hope that GMS adds some better explanation of the wet/dry flag to the GMS Help file- right now it seems to be missing from the LPF dialog discussion. Also GMS should not map it as zero as the default- it would be better to map it 0.5 or -0.5 as the default. Or better yet, have it assigned based on the rewetting boxes checked!

I also checked the box to eliminate the vertical leakance correction as I understand that can cause difficult non-linearities in the solution in rewetting areas, that can make things harder to settle.

Thanks again for the response.

Matthew,

There is probably nothing wrong with the solution. The apparent flooding is an artefact of how "flooded" cells are displayed in GMS. If the head in the uppermost active cell is above the top of the cell, it will be marked as flooded. If this is the top layer of the model, then you have real flooding. In the case of lower cells, however, the overlying cell has probably run dry and been inactivated, but heads have then risen above the top of the underlying active cell. You may need to allow re-wetting so that the cell above can become active again, but of course that may lead to instability.

If you want to show which areas are flooded/artesian, you would be better of using DATA, 3D Data>2D Data and saving "Highest active value in ij column" In the 2D DATA CALCULATOR menu, subtract the ground surface from this head, which will give you the head above ground surface. Then contour it showing only positive values.

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...

×
×
  • Create New...