How to pick a value for the nudging time scale

General scientific issues regarding ROMS

Moderators: arango, robertson

Post Reply
Message
Author
sonaljit.m
Posts: 43
Joined: Wed Nov 30, 2016 11:18 pm
Location: University of Massachusetts Dartmouth

How to pick a value for the nudging time scale

#1 Unread post by sonaljit.m »

Hello,
I am trying to apply nudging to the sea surface height and the scalar fields (temp/salinity), in a domain with 1 km grid size. The objective is to relax the SSH and temp/salinity fields to the climatology input, and allow the velocities to evolve accordingly.

My question is - How do I pick an appropriate magnitude for the nudging timescale? How much (in days) would be considered as strong nudging, and how much is weak?
Also, for the scalar fields, is nudging recommended throughout the whole domain? Or, should it be confined to the boundaries?

Thank you.
Sonaljit.

User avatar
kate
Posts: 4088
Joined: Wed Jul 02, 2003 5:29 pm
Location: CFOS/UAF, USA

Re: How to pick a value for the nudging time scale

#2 Unread post by kate »

In general, we want as little nudging as we can get away with. So you need to have a goal with your nudging, like getting rid of a specific boundary problem, for instance. If that's your case, there's no reason to nudge beyond a boundary band.

How strong? I use a linear ramp from 30 days at the boundary to infinite (no nudging) inside, over a a band of say 20 grid points. That has served us to do the job for T and S, also nudging velocities from a larger domain. I have to admit we haven't tried nudging only T and S.

Nudging to SSH may or may not be helpful. SSH is a secondary variable in the model, derived from lateral divergences and convergences of the barotropic flow.

sonaljit.m
Posts: 43
Joined: Wed Nov 30, 2016 11:18 pm
Location: University of Massachusetts Dartmouth

Re: How to pick a value for the nudging time scale

#3 Unread post by sonaljit.m »

Thanks Kate.

My objective is to simulate particle transport in the coastal areas, and for that I need the model to evolve the surrounding oceanic circulation by taking climatological inputs (from HYCOM). However, my concern is that the sharp bathymetry features give rise to spurious pressure gradient currents, that can deviate the flow in the long run. I am trying to mitigate this problem by smoothing the bathymetry, and using strong viscosities (UV_VIS4, MIX_GEO_UV). However, these measures can only reduce the problem but not completely eliminate it. Hence, I am considering to use nudging on the fields.

Post Reply