GMS: understanding flow budget

[GMS User]

I successfully build a steady-state MODFLOW model (4 layers) with:

-specified head boundary condition = 0.1 m constant

-specified flow boundary condition = 71 m3/d constant

-recharge

Then I add:

-pumping well (extraction) = -5 m3/d constant

I'm not sure to understand the flow budget compared to querying each elements.

In the attached picture you can see the model representation and the values I get from the flow budget and from querying each element.

Flow budget (calibrated model):

flow in the model: +100 = 0 (from head boundary) + 35 (from recharge) +65 (from flow boundary)

flow out from the model: -100 (from the 2 head boundary condition).

1st question: The flux boundary is set to 71 m3/d but in the model I get only 65 m3/d.

Why this difference? Water in equlibrium is not getting in the model?

Flow budget (calibrated model and pumping well):

flow in the model: +103 = 3 (from head boundary, due to well extraction) + 35 (from recharge) +65 (from flow boundary)

flow out from the model: -88 (from the 2 head boundary condition) and -15 from pumping well.

2st question: The well pumping rate is set to 5 m3/d but in the model I get 15 m3/d.

Why this difference?

Thank you in advance.

[Bill Dolinar]

I'm wondering if the WEL package flow rates in the MODFLOW model are correct. For the calibrated model if you go to the Well Package dialog (MODFLOW | Optional Packages | WEL - Well), select the Q column, press control-C to copy the values and then paste them into Excel or some other spreadsheet does the sum equal 71? If it doesn't, then maybe the conceptual model isn't getting mapped quite correctly.

For the model with the added well, do the added values in the Well Package dialog add up to 5?

Another possibility for the low WELL value in the calibrated model could be that some of the specified flow cells have gone dry (showing the red triangle symbol).

[GMS User]

Thank you for your reply.

You were right: the conceptual model is not mapped correctly due to intersecting an oblique boundary condition. I redraw the flow coverage line, map it to modflow, run the model and get the correct values. Great.

I checked the MODFLOW | Optional Packages | WEL - Well and I get 3 cells (Cell ID) with -5 m3/d.

The well coverage is mapped to use layer from 2 to 4 (3 layers of a 4 layers model). The flow rate is -5 m3/d. In the Flow Budget I get -5 (m3/d) x 3 (layers) = -15 m3/d.

Is it correct?

[Bill Dolinar]

So in the well package dialog there are three cells with a flow of -5 each? If that's the case then you probably have the well in the conceptual model set so it doesn't use a screen. When a well is in a coverage set to have multiple layers and has no screen, mapping to MODFLOW will add a BC with the specified value (-5) to each layer, which is likely is not what you want. It may be that you want to set the conceptual model to use a well screen. Some other options include duplicating a coverage with the well for each layer and setting the separate value for each layer, setting the value to 1/3 the value, or maybe it would make the most sense to use the MNW1 or MNW2 package.

[GMS User]

My model has fixed bottom elevation for layers 1, 2 and 3 (for inserting barrier and constant head boundary condition) and interpolated bottom elevation for layer 4.

I thought it was easier to assign a well to layers than to change screen bottom depth (as I've to move it in order to get the capture zone intersecting the contamination plume).

Anyway I agree it's better to set the well coverage to use screen as my layers don't have the same thickness.

Thank you.