search

MODFLOW-USGS / mt3d-usgs

MT3D-USGS Repository
23 stars 12 forks source link

Symetric flow solution; Asymetric transport solution #7

Open emorway-usgs opened 7 years ago

emorway-usgs commented 7 years ago

In the UZT test problems involving a vertical column (in particular the problem titled 'UZT_Disp_Lamb01_TVD'), the way mass is routed/removed in the saturated zone in the bottom 2 layers of the model appears to be asymmetric even though the flow solution shows symmetry. The percent discrepancy is low enough that it isn't a problem, however, when more time is available might be worth investigating further. In looking through the source code on 9/22/16 for roughly an hour (Vivek & Eric), nothing obvious stood out, will likely need to ferret out the issue by tracking values in the A matrix.

Notes for revisiting the probelm: monitor the concentrations in cells i,j,k=(1,1,210) and (1,3,210) for example, there is not reason the concentration should be different in these cells based on what is in the flow solution. Also, the concentrations will only be different when the slug of mass moving through the system arrives at depth.

emorway-usgs commented 7 years ago

This really only affected the TVD problem, mass discrepancies didn't seem to show up in the other test problems when the mass arrived at the lower boundary conditions.