Hi all, i am using ROMS3.3 and i am wondering if with this version i can add a passive tracer and study it's behavior using initial, and (horizontal/vertical) boundary conditions. (for example the same way as temperature).
Is it possible with this version? If yes, what would be the logical way or steps to do it?
If not, what would you suggest version wise speaking for such a simulation?
Thank you in advance!
Last edited by PhilipT on Fri Jun 12, 2015 7:07 pm, edited 1 time in total.
I expect that you can use passive tracers with that version - they've been around a while. I say try it and see. On the other hand, if you upgrade to the latest we'll have a better chance of answering specific questions.
So , i defined T_PASSIVE and ANA_BPFLUX,ANA_SPFLUX for zero surface and bottom fluxes.
Also, NPT==1 and Hout(inert)==T.
Insert the dye_01 at a specific location (zero everywhere and 10 to a certain location) in the init.nc and dye_(east,west,nort,south)_01=0 everywhere in the boundary file.
It seems to run pretty well but at some point after a year of run i get a huge error coming in the domain from the southern boundary giving huge negative and positive values for the tracer field.
You told us of your vertical boundary conditions on the passive tracer, but not the lateral boundary conditions. What are you using? For this, it might depend on the ROMS version - did you end up updating or going with your 3.3?
I still try with ROMS3.3.
I download the latest one also and working on that.
My lateral bry are all set to zero having the same dimension with temp and salt.
It behaves really good, but after a year when the concentration get low, a huge error comes in from the boundary where the tracer exits due to the circulation.
I have noticed that the problem starts exactly after one year.
Somehow it seems that the model needs to place something in, after one year from deployment!
There is something that i am missing in order to stop this behavior!
My lateral bry are all set to zero having the same dimension with temp and salt
which are open boundary values but what are your open boundary conditions?
Are they clamped, gradient, or radiation on the inert tracers?
If you update your code you will have much greater control on these settings, because you can set e.g. radiation/nudging on the active tracers (temp and salt) and clamp the passive dye (though radiation plus strong nudging to zero might be better).
With ROMS3.3 I think you are locked in to having the same conditions on all tracers.
John Wilkin: DMCS Rutgers University
71 Dudley Rd, New Brunswick, NJ 08901-8521, USA. ph: 609-630-0559 jwilkin@rutgers.edu
If you are having a problem after exactly one year, one might also ask exactly how you are setting the boundary values to zero. Is it a netcdf file? What is your calendar like - repeating after a year? Ending after a year? Show us the "ncdump -h" on it.