Roms nudging question (ocean_nud.nc and LnudgeTCLM)

[hdj002]

Hi. I am running ROMS 3.6, and am intrested in nudging with the climatology I have. I want to nudge the entire domain evenly, but I am still trying to grasp the nudging-settings.

The only nudging I have enabled with regard to nudging is:

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Logical switches (TRUE/FALSE) to read and process climatology fields.
! See glossary below for details.

LsshCLM == F ! sea-surface height
Lm2CLM == F ! 2D momentum
Lm3CLM == F ! 3D momentum

LtracerCLM == T T ! temperature, salinity, inert

! Logical switches (TRUE/FALSE) to nudge the desired climatology field(s).
! If not analytical climatology fields, users need to turn ON the logical
! switches above to process the fields from the climatology NetCDF file
! that are needed for nudging. See glossary below for details.

LnudgeM2CLM == F ! 2D momentum
LnudgeM3CLM == F ! 3D momentum

LnudgeTCLM == T T ! temperature, salinity, inert[/color]
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

I also have an ocean_nud.nc file. But what values should i choose for the variables (temp_NudgeCoef, salt_NudgeCoef and tracer_NudgeCoef)?

In the python script making the ocean_nud one can specify directions (west, east etc), and this will give values gradually declining towards 0, but I am interested in just an even nudging-field for the entire domain. What does it matter if i set the values to 1, vs 0.1? is it the strenght of the nudging, or is it related to the timescale?

Thanks in advance.

[kate]

First of all, I always avoid nudging if at all possible. I only use it to fix horrible boundary currents so having it at the edge only is exactly what I do. Why do you want to nudge everywhere?

I would run the python script once and see what sort of attributes it gives the nudging fields. Does it have units? They would be a timescale or an inverse timescale. Also, you could trace what ROMS does with the nudging fields once it reads them - does it take an inverse of them?

Then look at the Python script and figure out what it's doing. It should be pretty easy to make the nudging fields uniform instead of with the gradients.

[Ryan_Holmes]

If you want to nudge the whole domain with a single nudging time-scale then you can simply enable the cpp definition ANA_NUDGCOEF and set the nudging time scales using TNUDG in the .in file, avoiding the need for a nudging file ocean_nud.nc. Note that this method is not recommended for setting spatially dependent nudging parameter (see https://www.myroms.org/projects/src/ticket/627), but is fine for a constant value where you do not need to edit ana_nudgcoef.h.

[hdj002]

Thanks for the quick replies both of you!

The domain I am modelling is the arctic ocean, and our previous runs without nudging shows that the mixing of the atlantic current as it passes norway and svalbard dissipates way to fast because of vertical mixing caused by the steep bathymetry in the area. To try a different approach my supervisor suggested we try a nudging of first half a year, then 1 year, and see how this will turn out. the idea is to run the model from 2000-2015, with 4km resolution

About your idea Ryan_Holmes, it seems perfect for what I want. However, in my log file for a testrun I just did, ANA_NUDGECOEFF is described as:
ANA_NUDGCOEF Analytical spatially varying nudging time-scales.

But as long as I dont spesify any directonal nudging or anything, this should do a uniform spatially nudging over the entire domain?

[kate]

Ah well, we just need a better model. One that is less diffusive and one that scales better for these large problems.

[Ryan_Holmes]

Yes if you look in ana_nudgcoef.h you'll see that if no custom code is added it defaults to spatially uniform with the values specified in the .in file in TNUDG, M2NUDG and M3NUDG.