[MITgcm-support] How to calculate the internal tide disspation from MITgcm output ?

[Jody Klymak]

Hi Mike,

I don¡¯t know what You et al 2023 refers to ¨C do you have a journal name?

However your question is not very clear.  Eq 8 is a barotropic energy equation, and related the barotropic tide divergence to its terms: the conversion to baroclinic motions, C, and the barotropic dissipation \bar{D} and \bar{\epsilon}.  Note this equation is missing non-linear advection of the barotropic energy, but that is often small in deep water.

15-17 are probably baroclinic dissipations, though confusingly you have u¡¯ and u_{bc}.  However, there is of course no dissipation of potential energy in the barotropic equations.

C should balance the divergence of the energy flux and the dissipation in some volume, plus a non-linear baroclinic advection term.

\epsilon_{diff} is typically substantially less than \epsilon_{visc}, usually around 20%, though its of course possible to drive it to be different values based on how you choose your viscosities and diffusivities.  You have viscosity 10 times your diffusivity, so I wouldn¡¯t expect \epsilon_{visc} to be too high.

I always use Kang and Fringer for the energy decompositions (https://journals.ametsoc.org/view/journals/phoc/42/2/jpo-d-11-039.1.xml). They go through the derivation carefully as does Kang¡¯s thesis.

You may find my (outdated) energetics code useful:  https://github.com/jklymak/MITgcmcode

Cheers,   Jody

From: MITgcm-support <mitgcm-support-bounces at mitgcm.org> on behalf of Mike Fan <ifanliming at outlook.com>
Date: Tuesday, June 27, 2023 at 12:23 AM
To: MITgcm Support <mitgcm-support at mitgcm.org>
Subject: [MITgcm-support] How to calculate the internal tide disspation from MITgcm output ?
Äãͨ³£²»»áÊÕµ½À´×Ô ifanliming at outlook.com µÄµç×ÓÓʼþ¡£Á˽âÕâÒ»µãΪʲôºÜÖØÒª<https://aka.ms/LearnAboutSenderIdentification>
Hello, everyone

Recently I try to get the disspation of internal tide from MITgcm output, I have set the coefficients for constant viscosity, diffusion, and bottom friction. I want to know that, in this way, can I use the equations(15-18) of You et al. (2023) to do the job?

[cid:11fddd70-bb73-4fc8-a7ac-eade39c97c0c]
[cid:9631aa7a-893b-4d27-8daa-788aa288b2cc]

In my experience (in the only M2 internal tide model), the uncertainty of dissipation caused by diffusion (Equ.17) is the greatest. When the density perturbation (rho') is calculated by rho-mean(rho), this term (sum of Equ.(16-18) ~20GW) is absolutely dominant and is obviously larger than that estimated by Equ.8 (rhs ~6GW). When the density perturbation (rho') is calculated by harmonic analysis, the result (sum of Equ.(16-18) ~3GW) is smaller than that estimated by Equ.8. I'm not sure that if the choice of parameterization scheme may lead to differences in the calculation of dissipation, so I would like to ask, in a model with only M2 internal tide, how can I use the given constant parameters (viscosity, diffusion and bottom friction) to directly obtain the disspation generated by the internal tide?

Here is my "data" file setting:
# ====================
# | Model parameters |
# ====================
#
# Continuous equation parameters
 &PARM01
 TrefFile='Tannualref',
 sRef= 50*34.5,
 viscAh=5.0,
 viscAz=1.5E-4,
 diffKhT=5.0,
 diffKhS=5.0,
 diffKzT=1.5E-5,
 diffKzS=1.5E-5,
 no_slip_sides=.FALSE.,
 no_slip_bottom=.TRUE.,
 bottomDragQuadratic=2.5E-3,
 eosType='LINEAR',
 staggerTimeStep=.TRUE.,
 tAlpha=2.E-4,
 sBeta =0E-4,
 gravity=9.81,
 implicitFreeSurface=.TRUE.,
 exactConserv=.TRUE.
 nonHydrostatic=.FALSE.,
 useCoriolis=.TRUE.,
 hFacMin=0.2,
 hFacInf=0.2,
 hFacSup=1.8,
 tempStepping=.TRUE.,
 saltStepping=.FALSE.,
#- not safe to use globalFiles in multi-processors runs
#globalFiles=.TRUE.,
 readBinaryPrec=64,
 writeBinaryPrec=32,
 writeStatePrec=32,
 useSingleCpuIO=.TRUE.,
 &


Any reply or suggestion will be greatly appreciated.

-Mike
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.mitgcm.org/pipermail/mitgcm-support/attachments/20230627/da7c97e5/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 15-18.png
Type: image/png
Size: 101901 bytes
Desc: 15-18.png
URL: <http://mailman.mitgcm.org/pipermail/mitgcm-support/attachments/20230627/da7c97e5/attachment-0002.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 8.png
Type: image/png
Size: 17300 bytes
Desc: 8.png
URL: <http://mailman.mitgcm.org/pipermail/mitgcm-support/attachments/20230627/da7c97e5/attachment-0003.png>