[MITgcm-support] thanks martin

Martin Losch mlosch at awi-bremerhaven.de
Mon Oct 2 04:28:38 EDT 2006 

Hi Rima, I would just use the defaults (that is comment them out  
completely, the defaults are set in model/src/set_defaults.F)

M.
On 2 Oct 2006, at 09:40, Riema Rachmayani wrote:

> dear martin,
> you were right about lines in data:
> implicSurfPress=0.5,
> implicDiv2DFlow=0.5,
> i change them become 1.0 respectively....is that correct??
> thanks martin
>
> regards,
> rima, indonesia
>
> mitgcm-support-request at mitgcm.org wrote: Send MITgcm-support  
> mailing list submissions to
> mitgcm-support at mitgcm.org
>
> To subscribe or unsubscribe via the World Wide Web, visit
> http://mitgcm.org/mailman/listinfo/mitgcm-support
> or, via email, send a message with subject or body 'help' to
> mitgcm-support-request at mitgcm.org
>
> You can reach the person managing the list at
> mitgcm-support-owner at mitgcm.org
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of MITgcm-support digest..."
>
>
> Today's Topics:
>
> 1. question... (Riema Rachmayani)
> 2. Re: question... (Martin Losch)
> 3. Numerical diffusion (Lars Inge Enstad)
> 4. Re: Numerical diffusion (chris hill)
> 5. Re: Numerical diffusion (Martin Losch)
> 6. Re: Numerical diffusion (Baylor Fox-Kemper)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Fri, 29 Sep 2006 05:09:34 +0100 (BST)
> From: Riema Rachmayani
> Subject: [MITgcm-support] question...
> To: mitgcm-support at mitgcm.org
> Message-ID: <20060929040934.71651.qmail at web53801.mail.yahoo.com>
> Content-Type: text/plain; charset="iso-8859-1"
>
> dear all,
>
> i've already run verification of internal wave, i'm not change the  
> input...i just run it!!....when i check the input, Tref is various  
> to the vertical layer (20 layer), and Sref is uniform to the  
> vertical layer (20 layer), nonhydrostatic=.false....(it mean the  
> equation still hydrostatic right?)...and then i change  
> nonhydrostatic=.true. , when i run it again...there was a message :
>
> config_check : nonhydrostatic NOT SAFE with non fully implicit  
> barotopic solver
> config_check : to by-pass this STOP, comment this test and re- 
> compile config_check
> STOP ABNORMAL END : S/R Confid_check statement executed
>
> i thought that it happen because Sref uniform, so i change Sref  
> became various like Tref, but when i run it again, there was  
> appears the same message...
>
> what should i do??
>
> thx mitgcm
>
> best regard,
> rima, indonesia
>
>
> mitgcm-support-request at mitgcm.org wrote: Send MITgcm-support  
> mailing list submissions to
> mitgcm-support at mitgcm.org
>
> To subscribe or unsubscribe via the World Wide Web, visit
> http://mitgcm.org/mailman/listinfo/mitgcm-support
> or, via email, send a message with subject or body 'help' to
> mitgcm-support-request at mitgcm.org
>
> You can reach the person managing the list at
> mitgcm-support-owner at mitgcm.org
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of MITgcm-support digest..."
>
>
> Today's Topics:
>
> 1. Specifying the tile size in curvilinear grids (Ian G. Fenty)
> 2. Re: OBC problem: spurious boundary jets with C-D coupling
> (Mark Hadfield)
> 3. Exf depends on cal? (Mark Hadfield)
> 4. Re: Exf depends on cal? (Dimitris Menemenlis)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Wed, 27 Sep 2006 15:13:47 -0400
> From: "Ian G. Fenty"
> Subject: [MITgcm-support] Specifying the tile size in curvilinear
> grids
> To: mitgcm-support at mitgcm.org
> Message-ID: <451ACD6B.9020108 at mit.edu>
> Content-Type: text/plain; charset=ISO-8859-1; format=flowed
>
> Hello,
> Is there a way to pre-specify exactly the x,y size of each tile at run
> time even if this means they have non-uniform sizes? If so, would each
> tile then be known by a unique thread-id which could be known, or
> assigned, beforehand?
>
> Thanks,
> Ian Fenty
>
>
> ------------------------------
>
> Message: 2
> Date: Thu, 28 Sep 2006 09:49:14 +1200
> From: Mark Hadfield
> Subject: Re: [MITgcm-support] OBC problem: spurious boundary jets with
> C-D coupling
> To: mitgcm-support at mitgcm.org
> Message-ID: <451AF1DA.3060708 at niwa.co.nz>
> Content-Type: text/plain; charset="iso-8859-1"
>
> Thanks, Martin and Chris.
>
> chris hill wrote:
> > As Martin notes the code for CD with OBCS has not been written.
> > What resolution are you working at? Mostly CD is used for situations
> > where the deformation radius is not resolved i.e. resolutions >  
> 25km.
> > For a limited area run it may not be needed. For resolutions where
> > eddy scale is resolved or at least things are eddy permitting then
> > sub-grid dissipation is more cleanly (although its still a dirty
> > business) handled by either fixed laplacian and biharmonic  
> viscosities
> > or by the flow dependent smag/leith form (described in the online  
> docs).
> The simulation I am doing right now is at fairly coarse resolution (25
> km) and is diagnostic & linear, ie
>
> saltStepping = .FALSE.
> tempStepping = .FALSE.
> momStepping = .TRUE.
> momAdvection = .FALSE.
>
> Simulations of this sort are prone to developing grid-scale noise,  
> as I
> have found with ROMS, but biharmonic viscosity controls that nicely,
> thanks. Future, more realistic (I hope) simulations will use finer
> resolution and full dynamics.
>
> The decision to start with the CD scheme enabled wasn't a decision,
> really, just inertia: I started with one of the verification  
> experiments
> and started hacking.
>
> -- 
> Mark Hadfield "Kei puwaha te tai nei, Hoea tahi tatou"
> m.hadfield at niwa.co.nz
> National Institute for Water and Atmospheric Research (NIWA)
>
>
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: http://forge.csail.mit.edu/pipermail/mitgcm-support/ 
> attachments/20060928/c36d3d57/attachment-0001.html
>
> ------------------------------
>
> Message: 3
> Date: Thu, 28 Sep 2006 13:00:22 +1200
> From: Mark Hadfield
> Subject: [MITgcm-support] Exf depends on cal?
> To: MITgcm Support
> Message-ID: <451B1EA6.8050804 at niwa.co.nz>
> Content-Type: text/plain; charset=ISO-8859-1; format=flowed
>
> When I tried to build MItgcm with the exf package, but without the cal
> package, it failed during preprocessing with messages about missing
> headers that obviously belong to the cal package. (Or was it during
> linking? Whatever. It failed, the messages clearly implicated cal, and
> when I "cal" to packages.conf, it worked.)
>
> I presume, therefore, that pkg/pkg_depend should have an entry like
>
> exf +cal
>
> -- 
> Mark Hadfield "Kei puwaha te tai nei, Hoea tahi tatou"
> m.hadfield at niwa.co.nz
> National Institute for Water and Atmospheric Research (NIWA)
>
>
>
>
>
> ------------------------------
>
> Message: 4
> Date: Wed, 27 Sep 2006 21:23:11 -0700
> From: Dimitris Menemenlis
> Subject: Re: [MITgcm-support] Exf depends on cal?
> To: mitgcm-support at mitgcm.org
> Message-ID: <451B4E2F.4010608 at sbcglobal.net>
> Content-Type: text/plain; charset=ISO-8859-1; format=flowed
>
> Mark, you are right. pkg/exf requires pkg/cal.
> I have modified pkg/pkg_depend as suggested and checked it into CVS.
> Cheers, Dimitris
>
>
> ------------------------------
>
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support
>
>
> End of MITgcm-support Digest, Vol 39, Issue 29
> **********************************************
>
>
> Send instant messages to your online friends http:// 
> uk.messenger.yahoo.com
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: http://forge.csail.mit.edu/pipermail/mitgcm-support/ 
> attachments/20060929/0e0e7488/attachment-0001.html
>
> ------------------------------
>
> Message: 2
> Date: Fri, 29 Sep 2006 08:48:09 +0200
> From: Martin Losch
> Subject: Re: [MITgcm-support] question...
> To: MITgcm Support
> Message-ID:
> Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed
>
> Dear Rima,
>
> this check doesn't have anything to do with Tref or Sref.
> S/R (=subroutine) config_check.F tests the consistency of a few
> things. In this case the message is quite clear, isn't it:
> "nonhydrostatic NOT SAFE with non fully implicit barotopic solver".
>
> As far as I can see, this message is related to these lines in data:
> implicSurfPress=0.5,
> implicDiv2DFlow=0.5,
> which make the solver semi-implicit (I am guessing, I have never used
> these flags myself), they make the model use a Crank-Nickelson time
> stepping scheme (according to model/inc/PARAMS.h, where most
> parameters are briefly described).
>
> You have two options:
> 1. comment out these lines to go back to full implicit solver (which
> i would do, maybe someone can comment on WHY the solver is not fully
> implicit in this experiment)
> 2. Ignore the warning and comment out the corresponding STOP
> statement in model/src/check_config.F (as suggested in the second
> line of the warning, but it's probably not safe.)
>
> Martin, Germany
>
> PS. there is also a flag "implicitIntGravWave" (default=false), which
> you may want to try in conjunction with internal gravity wave .
>
> On Sep 29, 2006, at 6:09 AM, Riema Rachmayani wrote:
>
> > dear all,
> >
> > i've already run verification of internal wave, i'm not change the
> > input...i just run it!!....when i check the input, Tref is various
> > to the vertical layer (20 layer), and Sref is uniform to the
> > vertical layer (20 layer), nonhydrostatic=.false....(it mean the
> > equation still hydrostatic right?)...and then i change
> > nonhydrostatic=.true. , when i run it again...there was a message :
> >
> > config_check : nonhydrostatic NOT SAFE with non fully implicit
> > barotopic solver
> > config_check : to by-pass this STOP, comment this test and re-
> > compile config_check
> > STOP ABNORMAL END : S/R Confid_check statement executed
> >
> > i thought that it happen because Sref uniform, so i change Sref
> > became various like Tref, but when i run it again, there was
> > appears the same message...
> >
> > what should i do??
> >
> > thx mitgcm
> >
> > best regard,
> > rima, indonesia
>
>
> ------------------------------
>
> Message: 3
> Date: Fri, 29 Sep 2006 10:51:30 +0200
> From: Lars Inge Enstad
> Subject: [MITgcm-support] Numerical diffusion
> To: mitgcm-support at mitgcm.org
> Message-ID: <1159519890.3396.3.camel at fikentre.ii.uib.no>
> Content-Type: text/plain; charset=ISO-8859-15
>
> Hello
>
> I am currently trying to implement an alternative vertical turbulence
> model in MITgcm (GOTM (General Ocean Turbulence Model Burchard et
> al.), has this been tried before?). My question is about the numerical
> diffusion in the momentum advection scheme in MITgcm. How does this
> depend on the spatial and temporal discretisation? What is the  
> momentum
> advection scheme with the lowest possible numerical diffusion in
> MITgcm?
>
> One of the test cases I want to verify the code with is channel
> flow. In previous models I have used for channel flow, I have used a
> constant pressure gradient to force the flow. Is there some easy way
> to set up MITgcm this way?
>
> Best Regards
> Lars Inge Enstad
>
> -- 
> Lars Inge Enstad
> Post Doc
> Computational Mathematics Unit
> Bergen Center for Computational Science
> UNIFOB/University of Bergen
> Thormøhlensgt. 55
> N-5008 Bergen, Norway
>
>
>
>
> ------------------------------
>
> Message: 4
> Date: Fri, 29 Sep 2006 07:32:49 -0400
> From: chris hill
> Subject: Re: [MITgcm-support] Numerical diffusion
> To: mitgcm-support at mitgcm.org
> Message-ID: <451D0461.5000909 at mit.edu>
> Content-Type: text/plain; charset=ISO-8859-15; format=flowed
>
> Lars,
>
> The centered second order scheme is not diffusive, but it is  
> dispersive!
>
> There is a section in the online reference manual
>
> http://mitgcm.org/r2_web_testing/latest/online_documents/node85.html
>
> on how ( mostly horizontal ) viscosity is calculated. For vertical
> mixing there is Large et al KPP code incorporated into the dynamics.
>
> Martin Losch at AWI was recently thinking about whether to connect to
> the GOTM suite of schemes. He may have some thoughts on this. Martin
> also has a nice channel setup that he may be willing to share? He  
> is on
> this list.
>
> Chris
>
>
> Lars Inge Enstad wrote:
> > Hello
> >
> > I am currently trying to implement an alternative vertical  
> turbulence
> > model in MITgcm (GOTM (General Ocean Turbulence Model Burchard et
> > al.), has this been tried before?). My question is about the  
> numerical
> > diffusion in the momentum advection scheme in MITgcm. How does this
> > depend on the spatial and temporal discretisation? What is the  
> momentum
> > advection scheme with the lowest possible numerical diffusion in
> > MITgcm?
> >
> > One of the test cases I want to verify the code with is channel
> > flow. In previous models I have used for channel flow, I have used a
> > constant pressure gradient to force the flow. Is there some easy way
> > to set up MITgcm this way?
> >
> > Best Regards
> > Lars Inge Enstad
> >
>
>
>
> ------------------------------
>
> Message: 5
> Date: Fri, 29 Sep 2006 15:16:40 +0200
> From: Martin Losch
> Subject: Re: [MITgcm-support] Numerical diffusion
> To: mitgcm-support at mitgcm.org
> Message-ID: <9636ED5C-87E3-400C-8BD2-83D387AAB75A at awi-bremerhaven.de>
> Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed
>
> Please wait until Monday,
>
> M.
> On 29 Sep 2006, at 13:32, chris hill wrote:
>
> > Lars,
> >
> > The centered second order scheme is not diffusive, but it is
> > dispersive!
> >
> > There is a section in the online reference manual
> >
> > http://mitgcm.org/r2_web_testing/latest/online_documents/node85.html
> >
> > on how ( mostly horizontal ) viscosity is calculated. For vertical
> > mixing there is Large et al KPP code incorporated into the dynamics.
> >
> > Martin Losch at AWI was recently thinking about whether to connect
> > to the GOTM suite of schemes. He may have some thoughts on this.
> > Martin also has a nice channel setup that he may be willing to
> > share? He is on this list.
> >
> > Chris
> >
> >
> > Lars Inge Enstad wrote:
> >> Hello
> >> I am currently trying to implement an alternative vertical  
> turbulence
> >> model in MITgcm (GOTM (General Ocean Turbulence Model Burchard et
> >> al.), has this been tried before?). My question is about the
> >> numerical
> >> diffusion in the momentum advection scheme in MITgcm. How does this
> >> depend on the spatial and temporal discretisation? What is the
> >> momentum
> >> advection scheme with the lowest possible numerical diffusion in
> >> MITgcm? One of the test cases I want to verify the code with is
> >> channel
> >> flow. In previous models I have used for channel flow, I have  
> used a
> >> constant pressure gradient to force the flow. Is there some easy  
> way
> >> to set up MITgcm this way?
> >> Best Regards
> >> Lars Inge Enstad
> >
> > _______________________________________________
> > MITgcm-support mailing list
> > MITgcm-support at mitgcm.org
> > http://mitgcm.org/mailman/listinfo/mitgcm-support
>
>
>
> ------------------------------
>
> Message: 6
> Date: Fri, 29 Sep 2006 09:29:22 -0400
> From: Baylor Fox-Kemper
> Subject: Re: [MITgcm-support] Numerical diffusion
> To: mitgcm-support at mitgcm.org
> Message-ID:
> Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed
>
> Lars,
> If you are serious about including the GOTM, you should take a
> look at how KPP is implemented, MITgcm/pkg/kpp
> You may have significant difficulties in incorporating GOTM, as it
> seems that GOTM prescribes fluxes and changes to the mean flow not
> diffusivities/viscosities (which is how the MITgcm receives all
> current parameterizations). However, I would LOVE to see it
> implemented!
> -Baylor
>
> On Sep 29, 2006, at 7:32 AM, chris hill wrote:
>
> > Lars,
> >
> > The centered second order scheme is not diffusive, but it is
> > dispersive!
> >
> > There is a section in the online reference manual
> >
> > http://mitgcm.org/r2_web_testing/latest/online_documents/node85.html
> >
> > on how ( mostly horizontal ) viscosity is calculated. For vertical
> > mixing there is Large et al KPP code incorporated into the dynamics.
> >
> > Martin Losch at AWI was recently thinking about whether to connect
> > to the GOTM suite of schemes. He may have some thoughts on this.
> > Martin also has a nice channel setup that he may be willing to
> > share? He is on this list.
> >
> > Chris
> >
> >
> > Lars Inge Enstad wrote:
> >> Hello
> >> I am currently trying to implement an alternative vertical  
> turbulence
> >> model in MITgcm (GOTM (General Ocean Turbulence Model Burchard et
> >> al.), has this been tried before?). My question is about the
> >> numerical
> >> diffusion in the momentum advection scheme in MITgcm. How does this
> >> depend on the spatial and temporal discretisation? What is the
> >> momentum
> >> advection scheme with the lowest possible numerical diffusion in
> >> MITgcm? One of the test cases I want to verify the code with is
> >> channel
> >> flow. In previous models I have used for channel flow, I have  
> used a
> >> constant pressure gradient to force the flow. Is there some easy  
> way
> >> to set up MITgcm this way?
> >> Best Regards
> >> Lars Inge Enstad
> >
> > _______________________________________________
> > MITgcm-support mailing list
> > MITgcm-support at mitgcm.org
> > http://mitgcm.org/mailman/listinfo/mitgcm-support
>
>
>
> ------------------------------
>
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support
>
>
> End of MITgcm-support Digest, Vol 39, Issue 30
> **********************************************
>
> Send instant messages to your online friends http:// 
> uk.messenger.yahoo.com
>
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support

More information about the MITgcm-support mailing list