Code Comparison with similar mesh

Hi All,

For our Methane Hydrate code development project, now I am comparing OpenGeosys results with Code_Aster & Experimental Data . Our main purpose is to develop mechanical model for methane hydrate and to implement them in both OpenGeosys and Code_Aster. On the other hand, coupling performance of Code_Aster with TOUGH (HydrateResim) (internal and external coupling both) will be compared with the OpenGeosys considering the analogous modification in the both code (Code_Aster and OpenGeosys). The modification of THCM will depend on the Methane Hydrate project’s field data. The details of the project can be found here:

https://www.netl.doe.gov/research/oil-and-gas/methane-hydrates
https://www.netl.doe.gov/research/oil-and-gas/project-summaries/methane-hydrate
https://www.netl.doe.gov/research/labs-of-the-future/methane-hydrates
https://www.netl.doe.gov/research/oil-and-gas/methane-hydrates/gas-hydrate-global-assessment
https://www.netl.doe.gov/research/oil-and-gas/methane-hydrates/mh-06553hydrateprodtrial
https://www.netl.doe.gov/research/oil-and-gas/methane-hydrates/fire-in-the-ice

At present, OpenGeosys three elements are similar to Code_Aster (Quadrilateral, Hexahedron, and Triangular). In future if we need, any element from Code_Aster, I will implement them in OpenGeosys. I will also introduce similar modification for non-linear solver. After close observation I found, in both codes (OpenGeosys and Code_Aster), Liakopoulos test is common. To dig its bottom, I read following references

  1. Liakopoulos, A.C. (1965) “Transient flow through unsaturated porous media”, PhD Thesis, Civil Engineering, University of California.

  2. Narasimhan, T.N. and Witherspoon, P.A. (1978) “Numerical Model for saturated -unsaturated flow in deformable porous mdedia 3. Applications”, Water Resources Research, Vol 14, No 6, Page 1017-1034.

  3. Gawin, D., Baggio, P., and Schrefler, B.A. (1995) " Coupled heat, water and gas flow in deformable porous media", International Journal for Numerical Methods in Fluids, Vol. 20, 969-987.

  4. Gawin, D. and Schrefler, B.A. (1996) “Thermo-hydro-mechanical analysis of partially saturated porous materials”, Engineering Computations, Vol. 13 (7), 113-143.

  5. Gawin, D., Simoni, L. and Schrefler, B.A. (1997) “Numerical model for hydro-mechanical behaviour in deformable porous media: A benchmaark problem”, Computer Methods and Advance Geomechanics, Balkema, Rotterdam, Page: 1143-1148.

  6. Lewis, R.W. and Schrefler, B.A. (1998) “The Finite Element Method in the Static and Dynamic Deformationand Consolidation of Porous Media” Second Edition, John Wiler Sons, Page 167-174.

  7. Schrefler, B.A. and Scotta, R. (2001) “A fully coupled dynamic model for two-phase fluid flow in deformable porous mdedia” Computer methods in applied mechanics and engineering, 3223, 3223-3246.

  8. Granet, S. (2013) “WTNP124-Liakopoulos test case: Drainage of a column of water by the gravity force”, Code_Aster, V7.32.124, Revision: b2e73d847b87.

Now I have couple of questions:

a) Considering Triangular and Quadrilaeral element in xy and xz plane. It is found that only xz plane results are satisfactory. After generation of geometry, each and every time do I need to transfer xy plane data to xz plane? For your convenience, I attached data set for Quadrilateral element for both xy and xz plane.

b) In Code_Aster, the Liakopoulos case was simulated considering Hydrostatic Equilibrium condition considering gravity, while in OpenGeosys mechanical equilibrium condition is considered (to me more realistic !). I am wondering what is the procedure to obtain Hydrostatic Equilibrium condition in OpenGeosys? For your kind consideration, I attached Code_Aster Liakopoulos case in French and English (poor translation, sorry for inconvenience). I found that two files (*.ic and *. bc) need to modify to make it analogous to Code_Aster. Changing *.bc is straightforward and PRESSURE1 DIS_TYPE (CONSTANT 101325). To achieve, Hydrostatic condition in *.ic file for PRESSURE1, DIS_TYPE changed to GRADIENT. But the results are far beyond the expectation. Is there any other way?

In O.Kolditz et al. “Thermo-Hydro-Mechanical-Chemical Processes in Fractured Porous Media, Lecture notes in Computational Science and Engineering 86, Chapter 13 (Consolidation Process HnM Process)”, page 283 (Figure 13.13)& Page 284 (Figure 13.14)are closely related to my case. Though HM condition, but the mechanism of Hydrostatic condition in Figure 13.14 is very close.

c)In Code_Aster Mualem-Van-Genuchten model used, but in OpenGeosys for Liakopoulos case Brook-Corey function is used. Unfortunately, Van-Genuchten model is not working (may be due to my lack of knowledge). To change the model, I modified *.mmp file as follows to achieve Van-Genuchten

https://svn.ufz.de/ogs/wiki/public/doc-auto/by_ext/mmp/S_capillary_pressure

https://svn.ufz.de/ogs/wiki/public/doc-auto/by_ext/mmp/S_permeability_saturation

What is the appropriate way to obtain Van-Genuchten model? I am interested to compare both result.

d) Finally, one very basic question: what is the difference between “TWO_PHASE_FLOW” & “MULTI_PHASE_FLOW” ?

Thanks in advance.

Regards,

Mohammad

xz_xy plane.zip (1.2 MB)

v7.32.124_Eng.pdf (240 KB)

v7.32.124_French.pdf (263 KB)

Hi,

Now I have couple of questions:
a) Considering Triangular and Quadrilaeral element in xy and xz plane. It is found that only xz plane results are satisfactory. After generation of geometry, each and every time do I need to transfer xy plane data to xz plane? For your convenience, I attached data set for Quadrilateral element for both xy and xz plane.

In OGS5, you need a XZ mesh if gravity effects have to be included in your simulations. OGS6 doesn't have this constraint anymore.

b) In Code_Aster, the Liakopoulos case was simulated considering Hydrostatic Equilibrium condition considering gravity, while in OpenGeosys mechanical equilibrium condition is considered (to me more realistic !). I am wondering what is the procedure to obtain Hydrostatic Equilibrium condition in OpenGeosys? For your kind consideration, I attached Code_Aster Liakopoulos case in French and English (poor translation, sorry for inconvenience). I found that two files (*.ic and *. bc) need to modify to make it analogous to Code_Aster. Changing *.bc is straightforward and PRESSURE1 DIS\_TYPE \(CONSTANT 101325\)\. To achieve, Hydrostatic condition in \*\.ic file for PRESSURE1, DIS_TYPE changed to GRADIENT. But the results are far beyond the expectation. Is there any other way?

Using GRADIENT is a usual way to set hydrostatic condition. Alternatively you can use FUNCTION (a function of x,y,z) or DIRECT (manually specify values for every nodes) distribution types if you want. How are the results different from your expectation?

c)In Code_Aster Mualem-Van-Genuchten model used, but in OpenGeosys for Liakopoulos case Brook-Corey function is used. Unfortunately, Van-Genuchten model is not working (may be due to my lack of knowledge). To change the model, I modified *.mmp file as follows to achieve Van-Genuchten
https://svn.ufz.de/ogs/wiki/public/doc-auto/by_ext/mmp/S_capillary_pressure
https://svn.ufz.de/ogs/wiki/public/doc-auto/by_ext/mmp/S_permeability_saturation

What is the appropriate way to obtain Van-Genuchten model? I am interested to compare both result.

I cannot answer this because I'm not familiar with these stuff. Maybe you can take a look at the manual of OGS ver 4 (GeoSys/RockFlow). see section 15

There are couple of benchmark examples using van Genuchten model with MULTIPHASE_FLOW process. see e.g.

https://svn.ufz.de/ogs/browser/trunk/benchmarks/H2/LabGasInjec/H2_Permeability_GasPressure.mmp

d) Finally, one very basic question: what is the difference between "TWO_PHASE_FLOW" & "MULTI_PHASE_FLOW" ?

I'm not sure if TWO_PHASE_FLOW is still used, because there is no benchmarks using the process. To simulate multiphase flow, OGS5 provides two implementations which are different in terms of primary variables in the formulations:

- MULTI_PHASE_FLOW (pressure-pressure formulation)
- PS_GLOBAL (pressure-saturation formulation)

Best,
Nori

···

Thanks in advance.

Regards,
Mohammad

--
You received this message because you are subscribed to the Google Groups "ogs-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ogs-users+unsubscribe@googlegroups.com <mailto:ogs-users+unsubscribe@googlegroups.com>.
For more options, visit https://groups.google.com/d/optout\.

Hi Nori,

Thank you for your kind answer. I also want to thank Wenqing Wang.

In OGS5, you need a XZ mesh if gravity effects have to be included in
your simulations. OGS6 doesn’t have this constraint anymore.

I will move to OGS6 soon.

Using GRADIENT is a usual way to set hydrostatic condition.
Alternatively you can use FUNCTION (a function of x,y,z) or DIRECT
(manually specify values for every nodes) distribution types if you
want. How are the results different from your expectation?

Now, my hand calculation match with Hydrostatic condition. It is little tricky ! I have one question regarding FUNCTION. Is there any way to set FUNCTION along Z axis? Like, GeoSys/RockFlow Manual page 29, there is an option GRADIENT_Z (this key is no longer working!)

Van Genuchten (Same manual, page 66) for non-wetting cases are 14 and 15, but in source code it is 44. Now it is running, but not yet satisfied with the result. After comparison, I will let you know the update.

Regards,

Mohammad

···

On Tuesday, 23 May 2017 18:07:31 UTC-4, Norihiro Watanabe wrote:

Hi,

Now I have couple of questions:
a) Considering Triangular and Quadrilaeral element in xy and xz plane.
It is found that only xz plane results are satisfactory. After
generation of geometry, each and every time do I need to transfer xy
plane data to xz plane? For your convenience, I attached data set for
Quadrilateral element for both xy and xz plane.

In OGS5, you need a XZ mesh if gravity effects have to be included in
your simulations. OGS6 doesn’t have this constraint anymore.

b) In Code_Aster, the Liakopoulos case was simulated considering
Hydrostatic Equilibrium condition considering gravity, while in
OpenGeosys mechanical equilibrium condition is considered (to me more
realistic !). I am wondering what is the procedure to obtain Hydrostatic
Equilibrium condition in OpenGeosys? For your kind consideration, I
attached Code_Aster Liakopoulos case in French and English (poor
translation, sorry for inconvenience). I found that two files (*.ic and
*. bc) need to modify to make it analogous to Code_Aster. Changing *.bc
is straightforward and PRESSURE1 DIS_TYPE (CONSTANT 101325). To achieve, Hydrostatic condition in *.ic file for PRESSURE1, DIS_TYPE
changed to GRADIENT. But the results are far beyond the expectation. Is
there any other way?

Using GRADIENT is a usual way to set hydrostatic condition.
Alternatively you can use FUNCTION (a function of x,y,z) or DIRECT
(manually specify values for every nodes) distribution types if you
want. How are the results different from your expectation?

c)In Code_Aster Mualem-Van-Genuchten model used, but in OpenGeosys for
Liakopoulos case Brook-Corey function is used. Unfortunately,
Van-Genuchten model is not working (may be due to my lack of knowledge).
To change the model, I modified *.mmp file as follows to achieve
Van-Genuchten
https://svn.ufz.de/ogs/wiki/public/doc-auto/by_ext/mmp/S_capillary_pressure

https://svn.ufz.de/ogs/wiki/public/doc-auto/by_ext/mmp/S_permeability_saturation

What is the appropriate way to obtain Van-Genuchten model? I am
interested to compare both result.

I cannot answer this because I’m not familiar with these stuff. Maybe
you can take a look at the manual of OGS ver 4 (GeoSys/RockFlow). see
section 15

https://www.dropbox.com/s/bx9r21tv5m329sp/GSRF-manual.pdf?dl=0

There are couple of benchmark examples using van Genuchten model with
MULTIPHASE_FLOW process. see e.g.

https://svn.ufz.de/ogs/browser/trunk/benchmarks/H2/LabGasInjec/H2_Permeability_GasPressure.mmp

d) Finally, one very basic question: what is the difference between
“TWO_PHASE_FLOW” & “MULTI_PHASE_FLOW” ?

I’m not sure if TWO_PHASE_FLOW is still used, because there is no
benchmarks using the process. To simulate multiphase flow, OGS5 provides
two implementations which are different in terms of primary variables in
the formulations:

  • MULTI_PHASE_FLOW (pressure-pressure formulation)
  • PS_GLOBAL (pressure-saturation formulation)

Best,
Nori

Thanks in advance.

Regards,
Mohammad


You received this message because you are subscribed to the Google
Groups “ogs-users” group.
To unsubscribe from this group and stop receiving emails from it, send
an email to ogs-users+...@googlegroups.com

mailto:ogs-users+...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

On Tuesday, 23 May 2017 18:07:31 UTC-4, Norihiro Watanabe wrote:

Hi,

Now I have couple of questions:
a) Considering Triangular and Quadrilaeral element in xy and xz plane.
It is found that only xz plane results are satisfactory. After
generation of geometry, each and every time do I need to transfer xy
plane data to xz plane? For your convenience, I attached data set for
Quadrilateral element for both xy and xz plane.

In OGS5, you need a XZ mesh if gravity effects have to be included in
your simulations. OGS6 doesn’t have this constraint anymore.

b) In Code_Aster, the Liakopoulos case was simulated considering
Hydrostatic Equilibrium condition considering gravity, while in
OpenGeosys mechanical equilibrium condition is considered (to me more
realistic !). I am wondering what is the procedure to obtain Hydrostatic
Equilibrium condition in OpenGeosys? For your kind consideration, I
attached Code_Aster Liakopoulos case in French and English (poor
translation, sorry for inconvenience). I found that two files (*.ic and
*. bc) need to modify to make it analogous to Code_Aster. Changing *.bc
is straightforward and PRESSURE1 DIS_TYPE (CONSTANT 101325). To achieve, Hydrostatic condition in *.ic file for PRESSURE1, DIS_TYPE
changed to GRADIENT. But the results are far beyond the expectation. Is
there any other way?

Using GRADIENT is a usual way to set hydrostatic condition.
Alternatively you can use FUNCTION (a function of x,y,z) or DIRECT
(manually specify values for every nodes) distribution types if you
want. How are the results different from your expectation?

c)In Code_Aster Mualem-Van-Genuchten model used, but in OpenGeosys for
Liakopoulos case Brook-Corey function is used. Unfortunately,
Van-Genuchten model is not working (may be due to my lack of knowledge).
To change the model, I modified *.mmp file as follows to achieve
Van-Genuchten
https://svn.ufz.de/ogs/wiki/public/doc-auto/by_ext/mmp/S_capillary_pressure

https://svn.ufz.de/ogs/wiki/public/doc-auto/by_ext/mmp/S_permeability_saturation

What is the appropriate way to obtain Van-Genuchten model? I am
interested to compare both result.

I cannot answer this because I’m not familiar with these stuff. Maybe
you can take a look at the manual of OGS ver 4 (GeoSys/RockFlow). see
section 15

https://www.dropbox.com/s/bx9r21tv5m329sp/GSRF-manual.pdf?dl=0

There are couple of benchmark examples using van Genuchten model with
MULTIPHASE_FLOW process. see e.g.

https://svn.ufz.de/ogs/browser/trunk/benchmarks/H2/LabGasInjec/H2_Permeability_GasPressure.mmp

d) Finally, one very basic question: what is the difference between
“TWO_PHASE_FLOW” & “MULTI_PHASE_FLOW” ?

I’m not sure if TWO_PHASE_FLOW is still used, because there is no
benchmarks using the process. To simulate multiphase flow, OGS5 provides
two implementations which are different in terms of primary variables in
the formulations:

  • MULTI_PHASE_FLOW (pressure-pressure formulation)
  • PS_GLOBAL (pressure-saturation formulation)

Best,
Nori

Thanks in advance.

Regards,
Mohammad


You received this message because you are subscribed to the Google
Groups “ogs-users” group.
To unsubscribe from this group and stop receiving emails from it, send
an email to ogs-users+...@googlegroups.com

mailto:ogs-users+...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

hi,

below is an example of FUNCTION distribution type

$DIS_TYPE

FUNCTION

$GEO_TYPE

SUBDOMAIN

2 ; nr. of material groups below

0 1.693567e5+4000x±11559.45y

1 1.693567e5+4000x±11559.45y

The expression should be a+bx+cy+d*z, where a, b, c, and d are the constant to be input.

n

···

From: ogs-users@googlegroups.com ogs-users@googlegroups.com on behalf of Mohammad Islam nislamce@gmail.com

Sent: Thursday, May 25, 2017 5:15:44 AM

To: ogs-users

Subject: Re: [ogs-users] Code Comparison with similar mesh

Hi Nori,

Thank you for your kind answer. I also want to thank Wenqing Wang.

In OGS5, you need a XZ mesh if gravity effects have to be included in

your simulations. OGS6 doesn’t have this constraint anymore.

I will move to OGS6 soon.

Using GRADIENT is a usual way to set hydrostatic condition.

Alternatively you can use FUNCTION (a function of x,y,z) or DIRECT

(manually specify values for every nodes) distribution types if you

want. How are the results different from your expectation?

Now, my hand calculation match with Hydrostatic condition. It is little tricky ! I have one question regarding FUNCTION. Is there any way to set FUNCTION along Z axis? Like, GeoSys/RockFlow Manual page 29, there is an option GRADIENT_Z (this key is no longer working!)

Van Genuchten (Same manual, page 66) for non-wetting cases are 14 and 15, but in source code it is 44. Now it is running, but not yet satisfied with the result. After comparison, I will let you know the update.

Regards,

Mohammad

On Tuesday, 23 May 2017 18:07:31 UTC-4, Norihiro Watanabe wrote:

Hi,

Now I have couple of questions:

a) Considering Triangular and Quadrilaeral element in xy and xz plane.

It is found that only xz plane results are satisfactory. After

generation of geometry, each and every time do I need to transfer xy

plane data to xz plane? For your convenience, I attached data set for

Quadrilateral element for both xy and xz plane.

In OGS5, you need a XZ mesh if gravity effects have to be included in

your simulations. OGS6 doesn’t have this constraint anymore.

b) In Code_Aster, the Liakopoulos case was simulated considering

Hydrostatic Equilibrium condition considering gravity, while in

OpenGeosys mechanical equilibrium condition is considered (to me more

realistic !). I am wondering what is the procedure to obtain Hydrostatic

Equilibrium condition in OpenGeosys? For your kind consideration, I

attached Code_Aster Liakopoulos case in French and English (poor

translation, sorry for inconvenience). I found that two files (*.ic and

*. bc) need to modify to make it analogous to Code_Aster. Changing *.bc

is straightforward and PRESSURE1 $DIS_TYPE (CONSTANT 101325). To

achieve, Hydrostatic condition in *.ic file for PRESSURE1, $ DIS_TYPE

changed to GRADIENT. But the results are far beyond the expectation. Is

there any other way?

Using GRADIENT is a usual way to set hydrostatic condition.

Alternatively you can use FUNCTION (a function of x,y,z) or DIRECT

(manually specify values for every nodes) distribution types if you

want. How are the results different from your expectation?

c)In Code_Aster Mualem-Van-Genuchten model used, but in OpenGeosys for

Liakopoulos case Brook-Corey function is used. Unfortunately,

Van-Genuchten model is not working (may be due to my lack of knowledge).

To change the model, I modified *.mmp file as follows to achieve

Van-Genuchten


https://svn.ufz.de/ogs/wiki/public/doc-auto/by_ext/mmp/S_capillary_pressure


https://svn.ufz.de/ogs/wiki/public/doc-auto/by_ext/mmp/S_permeability_saturation

What is the appropriate way to obtain Van-Genuchten model? I am

interested to compare both result.

I cannot answer this because I’m not familiar with these stuff. Maybe

you can take a look at the manual of OGS ver 4 (GeoSys/RockFlow). see

section 15

https://www.dropbox.com/s/bx9r21tv5m329sp/GSRF-manual.pdf?dl=0

There are couple of benchmark examples using van Genuchten model with

MULTIPHASE_FLOW process. see e.g.

https://svn.ufz.de/ogs/browser/trunk/benchmarks/H2/LabGasInjec/H2_Permeability_GasPressure.mmp

d) Finally, one very basic question: what is the difference between

“TWO_PHASE_FLOW” & “MULTI_PHASE_FLOW” ?

I’m not sure if TWO_PHASE_FLOW is still used, because there is no

benchmarks using the process. To simulate multiphase flow, OGS5 provides

two implementations which are different in terms of primary variables in

the formulations:

  • MULTI_PHASE_FLOW (pressure-pressure formulation)

  • PS_GLOBAL (pressure-saturation formulation)

Best,

Nori

Thanks in advance.

Regards,

Mohammad

You received this message because you are subscribed to the Google

Groups “ogs-users” group.

To unsubscribe from this group and stop receiving emails from it, send

an email to ogs-users+...@googlegroups.com

mailto:ogs-users+...@googlegroups.com.

For more options, visit
https://groups.google.com/d/optout
.

On Tuesday, 23 May 2017 18:07:31 UTC-4, Norihiro Watanabe wrote:

Hi,

Now I have couple of questions:

a) Considering Triangular and Quadrilaeral element in xy and xz plane.

It is found that only xz plane results are satisfactory. After

generation of geometry, each and every time do I need to transfer xy

plane data to xz plane? For your convenience, I attached data set for

Quadrilateral element for both xy and xz plane.

In OGS5, you need a XZ mesh if gravity effects have to be included in

your simulations. OGS6 doesn’t have this constraint anymore.

b) In Code_Aster, the Liakopoulos case was simulated considering

Hydrostatic Equilibrium condition considering gravity, while in

OpenGeosys mechanical equilibrium condition is considered (to me more

realistic !). I am wondering what is the procedure to obtain Hydrostatic

Equilibrium condition in OpenGeosys? For your kind consideration, I

attached Code_Aster Liakopoulos case in French and English (poor

translation, sorry for inconvenience). I found that two files (*.ic and

*. bc) need to modify to make it analogous to Code_Aster. Changing *.bc

is straightforward and PRESSURE1 $DIS_TYPE (CONSTANT 101325). To

achieve, Hydrostatic condition in *.ic file for PRESSURE1, $ DIS_TYPE

changed to GRADIENT. But the results are far beyond the expectation. Is

there any other way?

Using GRADIENT is a usual way to set hydrostatic condition.

Alternatively you can use FUNCTION (a function of x,y,z) or DIRECT

(manually specify values for every nodes) distribution types if you

want. How are the results different from your expectation?

c)In Code_Aster Mualem-Van-Genuchten model used, but in OpenGeosys for

Liakopoulos case Brook-Corey function is used. Unfortunately,

Van-Genuchten model is not working (may be due to my lack of knowledge).

To change the model, I modified *.mmp file as follows to achieve

Van-Genuchten


https://svn.ufz.de/ogs/wiki/public/doc-auto/by_ext/mmp/S_capillary_pressure


https://svn.ufz.de/ogs/wiki/public/doc-auto/by_ext/mmp/S_permeability_saturation

What is the appropriate way to obtain Van-Genuchten model? I am

interested to compare both result.

I cannot answer this because I’m not familiar with these stuff. Maybe

you can take a look at the manual of OGS ver 4 (GeoSys/RockFlow). see

section 15

https://www.dropbox.com/s/bx9r21tv5m329sp/GSRF-manual.pdf?dl=0

There are couple of benchmark examples using van Genuchten model with

MULTIPHASE_FLOW process. see e.g.

https://svn.ufz.de/ogs/browser/trunk/benchmarks/H2/LabGasInjec/H2_Permeability_GasPressure.mmp

d) Finally, one very basic question: what is the difference between

“TWO_PHASE_FLOW” & “MULTI_PHASE_FLOW” ?

I’m not sure if TWO_PHASE_FLOW is still used, because there is no

benchmarks using the process. To simulate multiphase flow, OGS5 provides

two implementations which are different in terms of primary variables in

the formulations:

  • MULTI_PHASE_FLOW (pressure-pressure formulation)

  • PS_GLOBAL (pressure-saturation formulation)

Best,

Nori

Thanks in advance.

Regards,

Mohammad

You received this message because you are subscribed to the Google

Groups “ogs-users” group.

To unsubscribe from this group and stop receiving emails from it, send

an email to ogs-users+...@googlegroups.com

mailto:ogs-users+...@googlegroups.com.

For more options, visit
https://groups.google.com/d/optout
.

You received this message because you are subscribed to the Google Groups “ogs-users” group.

To unsubscribe from this group and stop receiving emails from it, send an email to ogs-users+unsubscribe@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.