Two diferent aquifers represented in the same layer

I would like to put the following questions:

I am modeling two aquifers separated by an impermeable layer (layer2),
a unconfined top aquifer (layer1) and a confined aquifer (layer3) in the bottom.

The unconfined top aquifer does not fill all area of model domain because my bottom confined aquifer outcrops in a small area and passes from a confined to a unconfined aquifer.

What is the most appropriated option to model this situation: 1) layer1 representing two aquifers, by assigning the respective hydraulic properties from each aquifer to each aquifer area or, 2) to decrease the thickness of layer1 when the bottom aquifer is outcropping?

I am modeling slice1 as free and movable, slice2 as fixed, slice3 as phreatic and slice4 as fixed.

If I use the first option, what is representing the piezometric distribution of layer1 in that area of my model that concerns the bottom aquifer?

(I am using FEFLOW v5.0)
Some help is appreciated.
Catarina

Dear Catarina,

I suggest that you try to understand better your concept, you need to be clear with hydrological relationship between all your geological layers.  If I understand correctly : your layer 2 and perhaps layer 3 could be unconfined and confined depending hydrogeological circumstancies, and your layer 1 and layer 2 are not present on all you surface?

Without drawings or scheme it is sometime difficult to be sure... :-\

Your layer 1 covers all your surface, but where it desappears, I suppose that the tickness of this layer is very small (10-15cm) (you could check it, it depends of your model design), then there you but the caracteristics of underlaying geological layer. It is the same with layer 2.

I suggest that you put free&movable for slice at the top, fixed for the bottom slice and unspecified for the rest. In the problem definition menu, check the help and I think you could understand and solve your last question.  It depends if your cells in the layer 1 could be dry and this depend of the geometry of model etc...

I hope this help you, feel free to ask more details  :)

Best regards

Dear Thierry
Thank you for your answer.

Yes, you have understated correctly the layer geometry.

I have put free&movable at the top, fixed, phreatic and fixed for the following slices.

Yes, I have problems with dry cells and I used the option “natural constrain conditions” (residual water depth for dry elements) to make the model converges. But the results for piezometry, after model runs, are too high in the dry cells area (they are set equal to the bottom elevation of layer1). So, I have read in the “help menu about constrains” that this constrain is optional and can be set to inoperative, then the hydraulic head can fall below the aquifer bottom. The results improved a lot with this.

But can I say that I have an unsaturated condition in this part of layer1? How can I be sure?

If this is correct then I found what I was looking for because I would not want a piezometric distribution where I have no aquifer.

The piezometric distribution for the confined aquifer is given in slice3.
Can I assume this piezometry represents confined and unconfined conditions in the same slice?

Can I please have a comment on this issue?
Best regards
Catarina

Dear Catarina



If you have dry cells in your layer 1, it is to say that your water level is under the bottom of the layer !

The result you obtain in this dry cells, is guided by your condition and is in your case (almost) at the bottom of the cells.

The question on unsaturated condition depend of what you understand with unsaturated condition ?

In the zone you have your dry cells, then your "true" water level is under the bottom of this layer then for you in layer 2 or layer 3.

your piezometry in layer 1 is available only for the saturated cells.  For a presentation of your results you have to take out the results in your dry cells.

One comment : perhaps your layer 3, in not always confined in all the area.

Your layer 3 could perhaps have confined and unconfined conditions it depends of what your are modelling?

But if you want to say that you see the piezometric level of layer 1 and layer 3 (you 2 aquifers), you have to take into account that in layer 1  your piezometric level could be a little bit higher due to the presence of semi-permeable, and If I suppose that layer is alimenting layer 3 etc...It depend of the situation you are modelling.


Hope this help you ! If you need more... :)


Have a nice day :D

Dear Thierry,
Thank for your comments.
You are correct in what you just said.
But still I have convergence problems.

The hydrogeological configuration is a plain area where I have a unconfined aquifer and a confined aquifer (separated by an impermeable layer) and a mountain area where my first aquifer disappears and I will just have the confined aquifer, that changes in this part to an unconfined aquifer.

In terms of layer geometry I suppose the best option is to reduce the thickness of layer1 and layer2 where they should not appear (in the hilly area). So, slice 2 and slice 3 elevations are quite high in this area, similar to the topography elevation. The piezoemtric head is always below those layer bottoms and I have convergence problems and dry cells.

Is there another way of creating this layer geometry in a way I will not have dry cells in the hilly zone?
If layers have to be continuous in the model domain how can I make then just disappear?
Some help is appreciated
Best regards
Catarina

Dear Catarina,

If you send me a drawing with a schematic cross section of you terrain (no need to put names) I could try to understand better and help you more accuratly.  With position of piezometric levels for each layer in the hilly zone and the flat zone and an idea of the altitude.

Have a nice WE  :)

OK!!
Very simply is something like this:
Elevation of topography from 0 (coast line) to 50 m (hilly zone)
Piezometric level in unconfined aquifer - from 0 to 10 m above sea level
Piezometry level in confined aquifer - from 0 to 8 m above sea level.
I have reduced layer1 and layer2 thickness in the hilly zone (so their elevation is something like 49 m and 48 m, above sea level, in this area)
Best regards,
Catarina

One solution
Perhaps you could design your model differently, for example try to change the bottom of layer 1 (and then your have to adapt layer 2, too probably) in order to have always the water leve in you first layer.  You have just to change the parameters when changing the lithology. 

I checked one of your previous message, I see that you put slice 2 (bottom of layer one) as fixed, I am not sure but for me it si not correct, because the layer 2 is then considered as confined and it seems to me that's not always true in your model?


have a nice day  :)

Dear Thierry
thank you for your suggestions.
The first model I have created was done as you said, just changing the hydraulic parameters in layer1. Then I would not have dry cells. Then the model calculates piezometric levels for layer1. Is it possible that this piezometric level is representing different aquifers in the same layer? I am not sure about this. That is why I have changed this approach.
Yes, you are right - layer2 is impermeable and not a confined aquifer. I have assigned it fixed because after a fixed slice there is always a confined aquifer, that is layer3.
Best regards.
Catarina

Just a small correction to my previous reply - layer3 is not always a confined aquifer, so slices2 and 3 should be unspecified (as you were saying).

If I consider the solution you refer in your last reply, then:
1) About my piezometric level, it is always in layer1, which is correct for a unconfined aquifer.
2) But for layer3 it is always above the top of the aquifer (elevations defined in 3D slice elevation) which is correct for a confined aquifer but not when this aquifer changes to unconfined.

Does this mean that I should only consider the piezometric level from layer1 in this area?
(values from all slices in this area are almost the same)
Best regards
Catarina

Quick answer
If the piezometric level is always upper than the top of the layer 3 (in your model) for me layer 3 is to be considered as confined.

The result on your layer 1 is partly available for your aquifer 1and for your aquifer 2.  Where your aquifer 2 is outcroping, you have normally few differences between results from layer 1, 2 and 3.  I think you could notice that at the change of your parameters in layer 1 (transition zone between aquifer 1- impermeable formation-aquifer 2) that you have some problems but it depends of the conductivity values.

in addition :

perhaps your convergence problem was due to others factors than dry cells? you find some questions hereafter to check some aspects of your model!

You said : "layer2 is impermeable and not a confined aquifer. I have assigned it fixed because after a fixed slice there is always a confined aquifer, that is layer3. "

If I understand correctly you assigned "fixed" to slice 2 (slice between layer 1 and layer 2)?

I was suprised that the water level of your aquifer 2 was below the water level of aquifer 1.  sometimes, It's difficult to understand thinks without maps !
The aquifer 2 is has an output under the sea? is there pumping ?
How do you put the boudary conditions?

are you running in steady state or transient flow?


Have a nice day

Dear Thierry,
For now I do not have convergence problems.
Some answers:
1) Initially yes, but I have made some changes. Now I have:
  Slice1 – free&movable
  Slice2 – unspecified
  Slice3 - unspecified
  Slice 4 – fixed
  Layer 1 – unconfined aquifer
  Layer 2 – impermeable layer
  Layer 3 – confined and unconfined aquifer
2) Both aquifers are discharging to a estuary zone not yet to the sea both. Saltwater edge is not considered.
3) I will have to put some pumping
4) Boundary conditions are:
North – Flux boundary (inflow)
East and west – impermeable
South - Head boundary (zero)
5) For now, still in steady sate.
Best regards,
Catarina

Hi, Catarina

I have similar questions with yours, For what I am modelling, there is also outcrop area of the bottom aquifers where confined changed to unconfined ,where my first unconfined aquifer also disappears, I wonder finally how you designed your model? Just by reducing the thickness of the layer 1 or 2 or? If so, how would you assign the precipitation recharge value (top/bottom inflow/outflow) to the outcrop area? Because in reality, the bottom aquifer will directly be recharged by rainfalls in the hilly area.
Is it possible if we establish a geological 3D model with other softwares and then plug into Feflow? For very complicated hydrogeological model? I wonder how Feflow solves these problems.

Thx a lot~