Help with Modflow Drain Package and model convergence?
modflow beginner
We are a group of senior engineering undergraduate students working
on a final design project. Our project is to lower the groundwater
table in at a landfill site to prevent contamination from the landfill
leaking down and entering the groundwater. We are using the Modflow
freeware Processing Modflow Version 5.3.1 to model the site. Nobody in
our group has used Modflow before, so we are just learning as we go.
If anyone could help us out with the problems we are having, we would
greatly appreciate it.
Before we model our design solutions, we have to model the existing
conditions at the landfill. We input all the data that we have into
the model (elevations, initial hydraulic heads, boundary conditions,
hydraulic conductivities, porosities). We have a line of fixed head
cells on the northern end of our model and a line of lower fixed head
cells on the southern end of the model as our boundaries. The other
borders of the model grid are no flow areas. We don’t have any
recharge inputted into our model. The model converges with this setup.
But we also have to model a drain which goes around three quarters of
the base of the landfill. When we put activate the drain in our model
and run it, the model no longer converges. Do you have any
suggestions?
Here are some questions that we have, if you could answer even a few
of them that would be great:
1. We have tried two different solvers SIP and PCG2 but they both
don’t converge even when we increase the number of iterations (they
basically start to converge but then reach a set of values for the
maximum head change that they just fluctuate between). We thought of
lowering our convergence criteria but it only converges when we set it
to ±1 ft, and we are not sure if this is accurate enough.
2. We are not sure of the conductance value for our drain. But the
engineer who gave us our project gave us a value for the flow out of
the drain that has been observed at the site. So we have been looking
at the water budget part of the output file for the drain and trying
to change the conductance of the drain cells in the model until the
flow out of the drain on the output file matches the flow observed at
the site. But when we set the conductance so that the drain is giving
us a flow that approximately matches the site conditions, the model
only converges to a maximum head change of ±1ft or higher.
3. When we run the model with the drain package activated, the water
budget says that the flow coming out of the drain is almost equal to
the flow coming out of our model. This is not the actual case at our
site, where the problem is that the drain is not able to capture a lot
of the groundwater flow. The water budget also says that the flow
coming out of the model is more than the flow coming in to the model
(there is around a 14% discrepancy between the values).
4. We also thought of looking at the contour lines for the hydraulic
heads in each layer and comparing them to the known site conditions.
But the engineer who provided the data to us does not know the actual
hydraulic head distribution precisely under the landfill. He gave us
the hydraulic heads upstream and downstream of the landfill but he
could only give us estimations of the approximate hydraulic head
distribution under the landfill. What is the usual way to tell if a
model is working? Through comparing the hydraulic head distribution
the model produces with the actual site conditions? By looking at the
model’s water budget and seeing if it matches the calculated water
budget for the site?
5. Some of our layers are discontinuous. We modelled these areas by
making the top elevation of that layer equal to the top elevation of
the layer below it so that it has no thickness. We also set these
cells with no thickness to inactive in our model using IBOUND. But we
have read some books that said that for discontinuous layers, you
should make the cell have some very small thickness and set the
hydraulic conductivities for that cell equal to the values of the cell
above or below it. What is the best way to model discontinuous
layers?
6. When we put in our drain, we used the cell by cell input method. In
the Modflow user’s manual we have, it mentioned the polyline input
method but our version does not seem to have this function (We have
Version 5.3.1, the freeware). Our drain is only about 5 ft wide, but
our cells are around 12 ft by 12 ft. Should we try refining the grid
to make the dimensions of each cell in the drain area around 5 ft
wide? Or is it ok to just vary the conductance until the flow out of
the drain shown in the water budget output file matches site
conditions?
Thank you for all your help.
Richard B. Winston
treat only part of the layer as representing the discontinuous horizon.
The remainder of the layer can be used to represent part of the horizon
above or below the discontinuous horizon. In the text below, each line
represents a model layer and each different number indicates which
horizon it is part of. Horizon 2 is discontinuous so in the places
where it is absent, horizon 1 is split between two different model
layers. There is no need to make the part of horizon 1 that is in layer
2 have a thickness of zero and personally, I would be leery of doing
that. I also would not make those cells inactive because that would
prevent any vertical flow between horizons 1 and 3.
11111111111
11111222222
33333333333
Another way is to use the Hydrogeologic Unit Flow (HUF) package.The HUF
package might not be supported by Processing Modflow. If I remember
correctly, your version of Processing Modflow is for MODFLOW-96 whereas
the HUF package was first introduced in MODFLOW-2000. If you wanted to
use HUF, you could consider switching from Processing Modflow to
ModelMuse as your graphical user interface for MODFLOW.
(http://water.usgs.gov/nrp/gwsoftware/ModelMuse/ModelMuse.html)
With regard to #6, you do not need to make the cells smaller just to
match the dimensions of the drain. The dimensions of the cell have no
effect on the amount of flow through the drain. The amount of flow is
determined by the difference in head between the cell and the drain and
the conductance of the drain. However, speaking of conductance, how are
you calculating the conductance? Be sure to read the section of the
MODFLOW manual where conductance is discussed. The dimensions of the
drain (not the cell) are used in calculating the conductance and that
calculation is done by the modeler outside of MODFLOW. While there is
no need to refine the grid to match the dimensions of the drain,
refining the grid, might (and might not) help with the model convergence.
With regard to model convergence, it sounds to me like too high a
conductance in the drains may be the cause of your convergence
problems. I suspect that what happens is that one one iteration the
water level is high enough for some of the drains to be active.
However, the flow out through the drains is so high that on the next
iteration, the water level drops below the bottom of the drain and they
shut off. That allows the water to rise above the drain elevation on the
next iteration. If that is the case a lower drain conductance might
help. You could also try varying the DAMP or RELAX parameters in the PCG
package or the ACCL parameter in the SIP package. Try looking at the
locations of the cells that PCG reports having the highest head change
or flux residual and see if those are drain cells.
With regard to #4, typical ways of assessing the quality of the model
would be to compare both the heads and fluxes to observed values. Heads
alone often do not have enough data to distinguish between models with
high flux and high hydraulic conductivities and models with low fluxes
and low hydraulic conductivities.
With regard to #3, the percent discrepancy in your model is far too high
to be acceptable. You should get it below 1%. Note however, that
although a percent discrepancy greater that 1% is a clear sign that a
model isn't working, a percent discrepancy less than 1% doesn't
guarantee the the model is OK.
modflow beginner wrote:
> Hi
> We are a group of senior engineering undergraduate students working
> on a final design project. Our project is to lower the groundwater
> table in at a landfill site to prevent contamination from the landfill
> leaking down and entering the groundwater. We are using the Modflow
> freeware Processing Modflow Version 5.3.1 to model the site. Nobody in
> our group has used Modflow before, so we are just learning as we go.
> If anyone could help us out with the problems we are having, we would
> greatly appreciate it.
>
> Before we model our design solutions, we have to model the existing
> conditions at the landfill. We input all the data that we have into
> the model (elevations, initial hydraulic heads, boundary conditions,
> hydraulic conductivities, porosities). We have a line of fixed head
> cells on the northern end of our model and a line of lower fixed head
> cells on the southern end of the model as our boundaries. The other
> borders of the model grid are no flow areas. We don�t have any
> recharge inputted into our model. The model converges with this setup.
> But we also have to model a drain which goes around three quarters of
> the base of the landfill. When we put activate the drain in our model
> and run it, the model no longer converges. Do you have any
> suggestions?
>
> Here are some questions that we have, if you could answer even a few
> of them that would be great:
>
> 1. We have tried two different solvers SIP and PCG2 but they both
> don�t converge even when we increase the number of iterations (they
> basically start to converge but then reach a set of values for the
> maximum head change that they just fluctuate between). We thought of
> lowering our convergence criteria but it only converges when we set it
> to �1 ft, and we are not sure if this is accurate enough.
>
> 2. We are not sure of the conductance value for our drain. But the
> engineer who gave us our project gave us a value for the flow out of
> the drain that has been observed at the site. So we have been looking
> at the water budget part of the output file for the drain and trying
> to change the conductance of the drain cells in the model until the
> flow out of the drain on the output file matches the flow observed at
> the site. But when we set the conductance so that the drain is giving
> us a flow that approximately matches the site conditions, the model
> only converges to a maximum head change of �1ft or higher.
>
> 3. When we run the model with the drain package activated, the water
> budget says that the flow coming out of the drain is almost equal to
> the flow coming out of our model. This is not the actual case at our
> site, where the problem is that the drain is not able to capture a lot
> of the groundwater flow. The water budget also says that the flow
> coming out of the model is more than the flow coming in to the model
> (there is around a 14% discrepancy between the values).
>
> 4. We also thought of looking at the contour lines for the hydraulic
> heads in each layer and comparing them to the known site conditions.
> But the engineer who provided the data to us does not know the actual
> hydraulic head distribution precisely under the landfill. He gave us
> the hydraulic heads upstream and downstream of the landfill but he
> could only give us estimations of the approximate hydraulic head
> distribution under the landfill. What is the usual way to tell if a
> model is working? Through comparing the hydraulic head distribution
> the model produces with the actual site conditions? By looking at the
> model�s water budget and seeing if it matches the calculated water
> budget for the site?
>
> 5. Some of our layers are discontinuous. We modelled these areas by
> making the top elevation of that layer equal to the top elevation of
> the layer below it so that it has no thickness. We also set these
> cells with no thickness to inactive in our model using IBOUND. But we
> have read some books that said that for discontinuous layers, you
> should make the cell have some very small thickness and set the
> hydraulic conductivities for that cell equal to the values of the cell
> above or below it. What is the best way to model discontinuous
> layers?
>
> 6. When we put in our drain, we used the cell by cell input method. In
> the Modflow user�s manual we have, it mentioned the polyline input
modflow beginner
Thanks for your help.
In the book we have on Modflow "3D Groundwater Modeling with
PMWIN" by Chiang and Kinzelbach, it says that the value of the drain
hydraulic conductance is usually unknown and must be adjusted during a
model calibration. This is why we thought we could vary the
conductance until the flow out of the drain matched our field data.
We also found a formula the drain conductance, which we
tried to use, that was in the EPA modflow tutorial:
C=KLW/M
where K=hydraulic conductivity of material around drain
(we have pea stone gravel, used 1 cm/s)
L=length of drain
W=width of drain (5 ft)
M=thickness of material surrounding drain
But this gives us a very high conductance (around 0.5 ft/s).
We weren't sure what to put as the length of the drain though. The
length of the drain for each cell we took as the length of the cell.
When we put in the drain cells, we used the cell-by-cell input method
(our version only has the cell-by-cell or zone input options). At the
actual physical site, the drain is in the shape of an elongated
semicircle. But when we put it in to the model we couldn't input
exactly a semicircle line for the drain since we put the drain in by
the cell by cell input method. I hope my explanation isn't too
confusing. Basically, we know the physical width of the cell and we
know that it makes a semicircle around the landfill site, but we don't
know how to deal with the length of the drain in the conductance
calculation.
Does the drain have to be all within one layer in a model? Our
model has five layers based on the cross section of the landfill we
were given(the landfill cap-which we were required to include in our
model even though there is no flow through it, a fill layer, an upper
clay layer, a till layer, and a lower clay layer). Right now our drain
is in the upper clay layer for some areas of the landfill and within
the fill layer in other areas. We have it set up like this because of
the drawings that were provided us, but some of the locations of the
drain are based on interpolating elevations so I think we could move
the drain so that the drain elevation is always within one of our
layers if necessary.
I looked at the output file for the solver and all of our
maximum head changes were occurring in our bottom clay layer (the
bottom layer, which has a very low permeability 10^(-7) ft/d). but the
maximum head changes occurred in different cells, not only the drain
cells.
I have changed our discontinuous layers so that their thickness
is 0.5 ft and they have the conductivity of the layer below them. Our
top fill layer is discontinuous. Also since we were told to include
the cap as a layer (the engineer wanted us to include it, in case we
had to model flow through it in the future due to leakance through the
landfill liner), the cap layer was discontinuous in some areas as
well. At first we had the cap cells as inactive since there was no
flow through them but we changed them to active and made their
thickness 0.5 ft in the areas where the cap doesn't exist.
When we run our model we sometimes get this error message: IN
SIP AT LAYER 3, ROW 79, COLUMN 65
THIS CAN OCCUR WHEN A CELL IS CONNECTED TO THE REST OF
THE MODEL THROUGH A SINGLE CONDUCTANCE BRANCH. CHECK
FOR THIS SITUATION AT THE INDICATED CELL."
I'm not sure what this means.
Thanks so much for all your help.
On Feb 27, 5:32 pm, "Richard B. Winston" <rbwins...@mindspring.com>
wrote:
> > borders of the model grid are no flow areas. We don t have any
> > recharge inputted into our model. The model converges with this setup.
> > But we also have to model a drain which goes around three quarters of
> > the base of the landfill. When we put activate the drain in our model
> > and run it, the model no longer converges. Do you have any
> > suggestions?
>
> > Here are some questions that we have, if you could answer even a few
> > of them that would be great:
>
> > 1. We have tried two different solvers SIP and PCG2 but they both
> > don t converge even when we increase the number of iterations (they
> > basically start to converge but then reach a set of values for the
> > maximum head change that they just fluctuate between). We thought of
> > lowering our convergence criteria but it only converges when we set it
> > to 1 ft, and we are not sure if this is accurate enough.
>
> > 2. We are not sure of the conductance value for our drain. But the
> > engineer who gave us our project gave us a value for the flow out of
> > the drain that has been observed at the site. So we have been looking
> > at the water budget part of the output file for the drain and trying
> > to change the conductance of the drain cells in the model until the
> > flow out of the drain on the output file matches the flow observed at
> > the site. But when we set the conductance so that the drain is giving
> > us a flow that approximately matches the site conditions, the model
> > only converges to a maximum head change of 1ft or higher.
>
> > 3. When we run the model with the drain package activated, the water
> > budget says that the flow coming out of the drain is almost equal to
> > the flow coming out of our model. This is not the actual case at our
> > site, where the problem is that the drain is not able to capture a lot
> > of the groundwater flow. The water budget also says that the flow
> > coming out of the model is more than the flow coming in to the model
> > (there is around a 14% discrepancy between the values).
>
> > 4. We also thought of looking at the contour lines for the hydraulic
> > heads in each layer and comparing them to the known site conditions.
> > But the engineer who provided the data to us does not know the actual
> > hydraulic head distribution precisely under the landfill. He gave us
> > the hydraulic heads upstream and downstream of the landfill but he
> > could only give us estimations of the approximate hydraulic head
> > distribution under the landfill. What is the usual way to tell if a
> > model is working? Through comparing the hydraulic head distribution
> > the model produces with the actual site conditions? By looking at the
> > model s water budget and seeing if it matches the calculated water
> > budget for the site?
>
> > 5. Some of our layers are discontinuous. We modelled these areas by
> > making the top elevation of that layer equal to the top elevation of
> > the layer below it so that it has no thickness. We also set these
> > cells with no thickness to inactive in our model using IBOUND. But we
> > have read some books that said that for discontinuous layers, you
> > should make the cell have some very small thickness and set the
> > hydraulic conductivities for that cell equal to the values of the cell
> > above or below it. What is the best way to model discontinuous
> > layers?
>
> > 6. When we put in our drain, we used the cell by cell input method. In
> > the Modflow user s manual we have, it mentioned the polyline input
> > method but our version does not seem to have this function (We have
> > Version 5.3.1, the freeware). Our drain is only about 5 ft wide, but
> > our cells are around 12 ft by 12 ft. Should we try refining the grid
> > to make the dimensions of each cell in the drain area around 5 ft
> > wide? Or is it ok to just vary the conductance until the flow out of
> > the drain shown in the water budget output file matches site
> > conditions?
>
> > Thank you for all your help.- Hide quoted text -
>
> - Show quoted text -
Richard B. Winston
the drain conductance rather than the hydraulic conductivity of the pea
gravel because the hydraulic conductivity of the aquifer will likely
limit the amount of water that can leave through the drain. The length
of the cell is probably a good first-order estimate of the length of the
drain required for the calculation.
Drains are not required to be all in one layer.
With regard to the error message from SIP, I think that means that one
or more cells are completely surrounded by inactive cells. A cell can
become inactive if the head in the cell drops below the bottom of the
layer. Only cells in convertible or unconfined layers can become inactive.
Rahim Alimohammadi
Dear
With profound regard
Rahim
Wen...@urscorp.com
I am building a groundwater model using Visual Modflow. The geological units dip about 40 degree to the South. The client wants the model to be constructed using the true geological layers. Can you please advise if there is a limit by MODFLOW on the dipping degree of the model layer?
Regards
Wen Yu
Principal Hydrogeologist
URS
Level 3, 20 Terrace Rd, East Perth, WA 6004, Australia
Tel: 61 8 9326 0213 Mobile: 0438688682 Fax: 61 8 9326 0296
mailto: Wen...@urscorp.com Visit our website at http://www.ap.urscorp.com
| This e-mail and any attachments contain URS Corporation confidential information that may be proprietary or privileged. If you receive this message in error or are not the intended recipient, you should not retain, distribute, disclose or use any of this information and you should destroy the e-mail and any attachments or copies. |
"Richard B. Winston" <rbwi...@mindspring.com>
Sent by: mod...@googlegroups.com 28/02/2010 06:32 AM
|
|
)
> borders of the model grid are no flow areas. We don’t have any
> recharge inputted into our model. The model converges with this setup.
> But we also have to model a drain which goes around three quarters of
> the base of the landfill. When we put activate the drain in our model
> and run it, the model no longer converges. Do you have any
> suggestions?
>
> Here are some questions that we have, if you could answer even a few
> of them that would be great:
>
> 1. We have tried two different solvers SIP and PCG2 but they both
> don’t converge even when we increase the number of iterations (they
> basically start to converge but then reach a set of values for the
> maximum head change that they just fluctuate between). We thought of
> lowering our convergence criteria but it only converges when we set it
> to ±1 ft, and we are not sure if this is accurate enough.
>
> 2. We are not sure of the conductance value for our drain. But the
> engineer who gave us our project gave us a value for the flow out of
> the drain that has been observed at the site. So we have been looking
> at the water budget part of the output file for the drain and trying
> to change the conductance of the drain cells in the model until the
> flow out of the drain on the output file matches the flow observed at
> the site. But when we set the conductance so that the drain is giving
> us a flow that approximately matches the site conditions, the model
> only converges to a maximum head change of ±1ft or higher.
>
> 3. When we run the model with the drain package activated, the water
> budget says that the flow coming out of the drain is almost equal to
> the flow coming out of our model. This is not the actual case at our
> site, where the problem is that the drain is not able to capture a lot
> of the groundwater flow. The water budget also says that the flow
> coming out of the model is more than the flow coming in to the model
> (there is around a 14% discrepancy between the values).
>
> 4. We also thought of looking at the contour lines for the hydraulic
> heads in each layer and comparing them to the known site conditions.
> But the engineer who provided the data to us does not know the actual
> hydraulic head distribution precisely under the landfill. He gave us
> the hydraulic heads upstream and downstream of the landfill but he
> could only give us estimations of the approximate hydraulic head
> distribution under the landfill. What is the usual way to tell if a
> model is working? Through comparing the hydraulic head distribution
> the model produces with the actual site conditions? By looking at the
> model’s water budget and seeing if it matches the calculated water
> budget for the site?
>
> 5. Some of our layers are discontinuous. We modelled these areas by
> making the top elevation of that layer equal to the top elevation of
> the layer below it so that it has no thickness. We also set these
> cells with no thickness to inactive in our model using IBOUND. But we
> have read some books that said that for discontinuous layers, you
> should make the cell have some very small thickness and set the
> hydraulic conductivities for that cell equal to the values of the cell
> above or below it. What is the best way to model discontinuous
> layers?
>
> 6. When we put in our drain, we used the cell by cell input method. In
> the Modflow user’s manual we have, it mentioned the polyline input
> method but our version does not seem to have this function (We have
> Version 5.3.1, the freeware). Our drain is only about 5 ft wide, but
> our cells are around 12 ft by 12 ft. Should we try refining the grid
> to make the dimensions of each cell in the drain area around 5 ft
> wide? Or is it ok to just vary the conductance until the flow out of
> the drain shown in the water budget output file matches site
> conditions?
>
> Thank you for all your help.
>
>
--
You received this message because you are subscribed to the Google Groups "MODFLOW Users Group" group.
To post to this group, send email to mod...@googlegroups.com.
To unsubscribe from this group, send email to modflow+u...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/modflow?hl=en.
Richard B. Winston
MODFLOW also does not account for the dip of beds when calculating the
conductance between cells. For a bed dipping 40 degrees, the
conductance might be 25% less (1 - COS(40 degrees)) if the dip were
taken into account. The uncertainty in hydraulic conductivity could
easily be more than this.