One layer to multilayer

[Chaachoua Mohamed]

Hello,

I have an hydrodynamic model elaborated with GMS 6.0, whose features are as follows:

1. The aquifer is unconfined

2. The model consists of a single layer

3. the model is simulated in transient of 35 years.

When I subdivided the model into several layers. I kept the first layer unconfined and the lowers confined (i think i should do so). For lowers layers I subdivised the specific yield by the the thickness of each layer to obtain the specific storage). The pumping is applicated through all the layers pro rata their thickness, I have considered the kz = Kh. But I do not get the same results of the original model.

How I could choose the parameters of my model multilayer among the original model to obtain the same results (the specific yield should become the specific storage for lower layers) and how to choose the vertical permeability of the layers?

I am very happy for any help

Thanks

[Sean Czarniecki]

You definitely do not want to make the lower layers confined. That will certainly make your results different from the original single-layer version.

[Chaachoua Mohamed]

It is not possible to consider the lower layers unconfined. When you subdivise the original layer to more than one layer you get for each layer a new top. This top is necessary above the head. So, each layer will be considered confined wich is more logical.

Thanks for your response and I am always waiting for your help

[Sean Czarniecki]

While you are correct conceptually, you need to call the layers "convertible" rather than "confined." The method for calculating the solution to each type of layer is different. Since you started with a single-layer unconfined model and then split the layers, the convertible option should be selected for each layer.

[Chaachoua Mohamed]

Hello Sean,

Thanks for your help. Of course I have considered the lower layers as convertible. I work with GMS 6.0 (lpf mode) so i have only two options : unconfined or convertible. The problem is can I choice parametres (S, Kz) from the original model and how can I reaffect pumping data through all layers so that I obtain the same results like my original model. I have explained in my first message how I have done it but unfortunately I could not obtain the same results.

Thanks for more help

[Sean Czarniecki]

The simpler parameter is Kz - In your single layer model, there is no restriction to water moving vertically. To get similar results when you split the layer, you need to make Kz = Kh for all layers. Otherwise, the preferred flow path is horizontal rather than vertical. Other people might be able to provide more information on specific storage/storativity, but I've found model results to be less sensitive to that parameter. I suggest selecting a ballpark value from literature based on your soil, then try a couple of sensitivity runs to see how much it impacts the results.

Good luck!

[Chaachoua Mohamed]

Thanks Sean for your promptness,

I'll try to make Specific storage=0 for lower convertible layers and the specific yield the same as the original model. Actually, the lower layers are not confined and under pression and the subdivision is just artificial.

Thanks for all