[MITgcm-support] diagnosing gmredi fluxes for tracers
Menemenlis, Dimitris (3248)
Dimitris.Menemenlis at jpl.nasa.gov
Thu Nov 14 10:16:57 EST 2013
Jean-Michel, could we add your description below to a file under MITgcn/doc or pkg/kpp? The information is too important and too tricky to live only in support list. Also it may require additions or adjustments later on.
Dimitris Menemenlis
818-625-6498
> On Nov 12, 2013, at 9:39 PM, Jean-Michel Campin <jmc at ocean.mit.edu> wrote:
>
> Hi Christoph,
>
> Does not look like you got an answer on support to your question after
> this clarification (including data.gmredi & GMREDI_OPTIONS.h).
> so if it's not too late, I will try to address your question.
>
> In your data.gmredi, which contains this comment:
> # GM_AdvForm : turn on GM Advective form (default=Skew flux form)
> GM_AdvForm is not set to TRUE, which means that you are using
> the Skew-Flux form of GM.
> You can find some description in the manual about this part,
> in section "6.4.1.3 Griffies Skew Flux".
>
> Now regarding pkg/diagnostics options:
> The isopycnal diffusion (Redi part) contribution is put into the
> diffusive flux, splitted into the 3 directions and,
> if using implicitDiffusion=.TRUE., with a distinction
> between explicit vertical flux (which contains the contribution
> Kwx d.T/dx + Kwy d/T/dy) and the implicit vertical flux
> (which contains the contribution Kwz d.T/dz).
> For instance, for temperature (but it's available for any tracer):
> 112 |DFrE_TH | 15 | |WM LR|degC.m^3/s |Vertical Diffusive Flux of Pot.Temperature (Explicit part)
> 113 |DFxE_TH | 15 | 114 |UU MR|degC.m^3/s |Zonal Diffusive Flux of Pot.Temperature
> 114 |DFyE_TH | 15 | 113 |VV MR|degC.m^3/s |Meridional Diffusive Flux of Pot.Temperature
> 115 |DFrI_TH | 15 | |WM LR|degC.m^3/s |Vertical Diffusive Flux of Pot.Temperature (Implicit part)
> And if using the Skew-Flux form, the GM part (which double Kwx & Kwy
> and cancel Kuz and Kvz if K_GM = K_iso ) will also fall into these diffusive
> flux diagnostics.
>
> With the advective form of GM ( GM_AdvForm=.TRUE.), and assuming the default
> GM_AdvSeparate=.FALSE. is used, the eulerian mean and the bolus transport
> are combined together to form the residual mean which is used to advect all
> tracers. And in this case, the GM contribution to the tracer tendency is
> part of the advective flux diagnostics, e.g. for temperature:
> 109 |ADVr_TH | 15 | |WM LR|degC.m^3/s |Vertical Advective Flux of Pot.Temperature
> 110 |ADVx_TH | 15 | 111 |UU MR|degC.m^3/s |Zonal Advective Flux of Pot.Temperature
> 111 |ADVy_TH | 15 | 110 |VV MR|degC.m^3/s |Meridional Advective Flux of Pot.Temperature
> Note that these diagnostics match exactly what the model computes
> since they come directly out of the model advection subroutines.
> But no distinction is made between the advective flux due to the eulerian mean
> and the one due to the bolus transport (hard to define anyway when
> the advection scheme is non-linear regarding the velocity).
>
> For this reason, but only available for temperature (sorry), there is an
> approximate estimation of the Bolus transport of temperature
> (consistent with a centered in time and space advection scheme, like tempAdvScheme=2)
> which is diagnose (in pkg/gmredi/gmredi_xtransport.F & pkg/gmredi/gmredi_ytransport.F)
> and available with the advective form:
> 209 |GM_ubT | 15 | 210 |UUr MR|degC.m^3/s |Zonal Mass-Weight Bolus Transp of Pot Temp
> 210 |GM_vbT | 15 | 209 |VVr MR|degC.m^3/s |Meridional Mass-Weight Bolus Transp of Pot Temp
>
> An other alternative, since all Redi (+GM if skewFlux form) matrix coefficient are available
> as diagnostics:
> 197 |GM_Kux | 15 | 198 |UU P MR|m^2/s |K_11 element (U.point, X.dir) of GM-Redi tensor
> 198 |GM_Kvy | 15 | 197 |VV P MR|m^2/s |K_22 element (V.point, Y.dir) of GM-Redi tensor
> 199 |GM_Kuz | 15 | 200 |UU MR|m^2/s |K_13 element (U.point, Z.dir) of GM-Redi tensor
> 200 |GM_Kvz | 15 | 199 |VV MR|m^2/s |K_23 element (V.point, Z.dir) of GM-Redi tensor
> 201 |GM_Kwx | 15 | 202 |UM LR|m^2/s |K_31 element (W.point, X.dir) of GM-Redi tensor
> 202 |GM_Kwy | 15 | 201 |VM LR|m^2/s |K_32 element (W.point, Y.dir) of GM-Redi tensor
> 203 |GM_Kwz | 15 | |WM P LR|m^2/s |K_33 element (W.point, Z.dir) of GM-Redi tensor
> would be to calculate (offline) an approximate evaluation (neglecting time correlation between tracer
> gradient and Redi-GM matrix coeff) of gmredi flux, base on the time-mean tracer gradient and
> the time-mean GM-Redi coefficient.
>
> Note that, but again for temperature only, some gmredi flux are available as diagnostics:
> 206 |GM_KuzTz| 15 | 207 |UU MR|degC.m^3/s |Redi Off-diagonal Temperature flux: X component
> 207 |GM_KvzTz| 15 | 206 |VV MR|degC.m^3/s |Redi Off-diagonal Temperature flux: Y component
> 208 |GM_KwzTz| 15 | |WM LR|degC.m^3/s |Redi main-diagonal vertical Temperature flux
>
> Hope I did not add too much confusion here.
> Cheers,
> Jean-Michel
>
>> On Wed, Oct 30, 2013 at 08:35:48AM +0100, Christoph Voelker wrote:
>> Hi Jean-Michel,
>>
>> I am indeed using implicitDiffusion=.TRUE. and my data.gmredi contains
>> that I am using the Large, Damabasoglu, Doney tapering scheme, and the
>> Visback variable kappa. I have attached data.gmredi and
>> GMREDI_OPTIONS.h, so if you could have a quick look I'd be grateful!
>>
>> Cheers, Christoph
>>
>>> On 10/29/13 11:05 PM, Jean-Michel Campin wrote:
>>> Hi Christoph,
>>>
>>> There are ways to get diagnostics for some of the different pieces of GM-Redi.
>>> But it would help if we know what parameter you are using (from data.gmredi)
>>> and if you are using implicitDiffusion=.TRUE., (from main paramter file "data").
>>>
>>> The fact that you have:
>>>> #define GM_BOLUS_ADVEC
>>>> in GMREDI_OPTIONS.h.
>>> does not mean that you are using the advective form (it's just that the
>>> code is compiled, and you can use it or not).
>>>
>>> Cheers,
>>> Jean-Michel
>>>
>>>> On Mon, Oct 28, 2013 at 11:34:05PM +0100, Christoph Voelker wrote:
>>>> Dear MITgcm'ers,
>>>>
>>>> I have a run of a biogeochemical model (i.e. with several additional
>>>> passive tracers) and I am trying to diagnose (and hopefully understand)
>>>> what the GMREDI eddy parameterization contributes to my tracer fluxes.
>>>> In doing so I got somewhat confused my the many options and parameter
>>>> settings that are possible in that package; I found it hard to follow
>>>> how the advective and diffusive fluxes are really calculated.
>>>>
>>>> My settings include
>>>> #define GM_VISBECK_VARIABLE_K
>>>> #define GM_BOLUS_ADVEC
>>>> in GMREDI_OPTIONS.h.
>>>> so, I am using the bolus velocity form, not the skew flux form, with
>>>> variable kappa_GM, a la Visbeck et al. '97. In addition I am using the
>>>> Large, Danabasoglu, Doney tapering scheme.
>>>>
>>>> My questions concern the diagnostics associated with the tracer fluxes.
>>>>
>>>> - Firstly, I assume that the explicit diffusive terms like DFrETr01
>>>> contain the Redi isopycnal diffusion, perhaps plus the additional
>>>> diffusion induced e.g. by the KPP sheme that I am using. Is this right?
>>>>
>>>> - Secondly, is the bolus advective flux part of the advective flux
>>>> diagnostics, like ADVrTr01? Or is this only the non-bolus advection?
>>>>
>>>> - And finally, is there a way to diagnose the advective tracer fluxes
>>>> from the bolus velocity alone, i.e. separate it from the background
>>>> velocity advection, in a diagnostic variable, or does one have to write
>>>> out the bolus stream function terms GM_PsiX and GM_PsiY, and then
>>>> compute the induced additional advection offline?
>>>>
>>>> Sorry if my questions are daft, but I really first tried to understand
>>>> all this from the code,
>>>>
>>>> Thanks, Christoph
>>>>
>>>> --
>>>> Christoph Voelker
>>>> Alfred Wegener Institute for Polar and Marine Research
>>>> Am Handelshafen 12
>>>> 27570 Bremerhaven, Germany
>>>> e: Christoph.Voelker at awi.de
>>>> t: +49 471 4831 1848
>>>>
>>>>
>>>> _______________________________________________
>>>> 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
>>
>>
>> --
>> Christoph Voelker
>> Alfred Wegener Institute for Polar and Marine Research
>> Am Handelshafen 12
>> 27570 Bremerhaven, Germany
>> e: Christoph.Voelker at awi.de
>> t: +49 471 4831 1848
>
>
>
>> _______________________________________________
>> 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
More information about the MITgcm-support
mailing list