How to setup the river discharges with only salinity but no temperature?

[yz966445]

Hi everyone,
Could anyone tell me how to setup the river discharge in ROMS model as I need?
I want the added river discharge affect the water salinity but no affecting the ocean temperature.

I am using "define ANA_PSOURCE" and "ana_psource.h" file to add the river discharge to my model.
For my own understanding, I setup the LtracerSrc in .in file as ''LtracerSrc == F T F ! temperature, salinity, inert".
However, I found the added river discharge still has lower temperature than the surrounding water.
Could anyone help me with that problem?


Thanks,
Yi Sui

[kate]

It doesn't make sense numerically to do what you want. As you see, the thing will become unstable. Lacking river temperatures, some have used air temperature instead.

[yz966445]

Hi Kate,
Thanks for you reply.
Could we do that setup? Let's setup the river water has the salinity = 0.4 psu and has the temperature as the surrounding water.
I think the default setup in ana_psource.h is
SOURCES(ng)%Tsrc(is,k,itemp)=5
SOURCES(ng)%Tsrc(is,k,isalt)=0.4

Do we must setup Tsrc(is,k,itemp) a certain value?

Yi Sui

It doesn't make sense numerically to do what you want. As you see, the thing will become unstable. Lacking river temperatures, some have used air temperature instead.

[kate]

For numerical stability you must give it something other than the water temperature at that grid point. Using the water temperature there is an unconditionally unstable numerical scheme - downwind advection. In MOM6, the default is to use air temperature instead, something you'd have anyway if you were using bulk fluxes. In ROMS, you'd have to code that in since it isn't the default.

[wilkin]

When I recoded the LwSrc option to introduce point sources as a divergence at the cell centers (instead of the the cell faces with LuvSrc) I did a sequence of tests with a modified RIVERPLUME1 test case and LtracerSrc == F T worked perfectly.

When LTracerSrc is false, tracer mass is added to compensate for the added volume to keep the receiving cell concentration unchanged. In the case of temperature, this is effectively adding heat. The relevant code is in step3d_t.F.

[yz966445]

Hi Wilkin,

Could you please tell me more information about how you setup? I want to make the LtracerSrc == F T worked perfectly. I am not sure how to code as you say in proper way. DId you recoded the LwSrc section? I very appreciate your help.

Yi Sui

When I recoded the LwSrc option to introduce point sources as a divergence at the cell centers (instead of the the cell faces with LuvSrc) I did a sequence of tests with a modified RIVERPLUME1 test case and LtracerSrc == F T worked perfectly.

When LTracerSrc is false, tracer mass is added to compensate for the added volume to keep the receiving cell concentration unchanged. In the case of temperature, this is effectively adding heat. The relevant code is in step3d_t.F.

[wilkin]

Could you please tell me more information about how you setup? I want to make the LtracerSrc == F T worked perfectly. I am not sure how to code as you say in proper way. DId you recoded the LwSrc section? I very appreciate your help.

The code update is documented in this ticket from July 2020:
https://www.myroms.org/projects/src/ticket/860
If you are using an earlier version of the code, LwSrc will give you wrong results.

The distinction between LuvSrc and LwSrc is documented in WikiROMS:
https://www.myroms.org/wiki/River_Runoff

All you have to do is make sure you have the correct river_Xposition and river_Eposition (i,j position of the rho-cell into which you want the river to flow, in ROMS FORTRAN index convention counting from 0,0 not 1,1)