inert passive tracer boundary not being read

General scientific issues regarding ROMS

Moderators: arango, robertson

Post Reply
Message
Author
cschultz
Posts: 13
Joined: Wed May 27, 2009 10:43 pm
Location: WHOI/MIT

inert passive tracer boundary not being read

#1 Post by cschultz » Fri Oct 24, 2014 10:16 pm

Hi all,

I'm having some trouble getting the boundaries for an inert passive tracer to work. I defined T_PASSIVE and ANA_BPFLUX, and changed NPT = 1 in the .in file. I provided initial conditions, dye_01_sflux and boundary conditions for the tracer. The initial condition and boundary are included in in the ic and bry netcdfs with all the other variables, and the surface flux is in a file of its own.

However, on a test I did with ANA_SPFLUX undefined, the dye is all gone within a few months. I can't get the model to even begin reading my boundary conditions for dye. Has anyone encountered this problem before? Am I missing something?

Thanks!

User avatar
kate
Posts: 3796
Joined: Wed Jul 02, 2003 5:29 pm
Location: CFOS/UAF, USA

Re: inert passive tracer boundary not being read

#2 Post by kate » Fri Oct 24, 2014 10:46 pm

What are the LBC values for your tracers? Four dye tracers with no boundary conditions would look like:

Code: Select all

   LBC(isTvar) ==   RadNud  RadNud  RadNud  RadNud \    ! temperature
                    RadNud  RadNud  RadNud  RadNud \    ! salinity
                    Clo     Clo     Clo     Clo    \    ! inert(1)
                    Clo     Clo     Clo     Clo    \    ! inert(2)
                    Clo     Clo     Clo     Clo    \    ! inert(3)
                    Clo     Clo     Clo     Clo         ! inert(4)
It seems you want RadNud instead of Clo.

As for the surface forcing, is that being read?

cschultz
Posts: 13
Joined: Wed May 27, 2009 10:43 pm
Location: WHOI/MIT

Re: inert passive tracer boundary not being read

#3 Post by cschultz » Fri Nov 14, 2014 8:26 pm

Hi Kate,

I fixed my .in and the dye data is read, both at the boundary and at the surface. However, in both experiments (I ran boundary input and surface input separately), the concentrations get really unreasonable really fast. I am posting the average dye concentration at 5 and 300m for the 7th month of my simulation, with dye initialized with 0 concentration on the domain, and a boundary input of 0.1 at the bottom layer. I am running this simulation with ice, and the only passive tracer option activated is T_PASSIVE. How can I stop the values from going negative? The 2 open boundaries were both defined with RadNud for dye.

Thank you!
Attachments
dye_300m.png
dye_5m.png

User avatar
kate
Posts: 3796
Joined: Wed Jul 02, 2003 5:29 pm
Location: CFOS/UAF, USA

Re: inert passive tracer boundary not being read

#4 Post by kate » Fri Nov 14, 2014 8:53 pm

Have you tried clamped dye boundary conditions? It looks like something is really wrong. If the dye BC is 0.1 at the bottom, is it 0.0 for the rest? I'd look much closer to the beginning time to try to figure it out.

User avatar
arango
Site Admin
Posts: 1128
Joined: Wed Feb 26, 2003 4:41 pm
Location: IMCS, Rutgers University
Contact:

Re: inert passive tracer boundary not being read

#5 Post by arango » Fri Nov 14, 2014 9:54 pm

Yes, I know. I has an update for it. Mark Hadfield put a post about this and the solution.

cschultz
Posts: 13
Joined: Wed May 27, 2009 10:43 pm
Location: WHOI/MIT

Re: inert passive tracer boundary not being read

#6 Post by cschultz » Tue Nov 18, 2014 5:14 pm

Thanks for the reply. I can't find the post with the solution, though, can anyone send me the link? Thanks again!

rtopper
Posts: 24
Joined: Fri Jul 04, 2014 2:32 pm
Location: MARUM, University of Bremen

Re: inert passive tracer boundary not being read

#7 Post by rtopper » Wed Nov 19, 2014 5:33 pm


User avatar
arango
Site Admin
Posts: 1128
Joined: Wed Feb 26, 2003 4:41 pm
Location: IMCS, Rutgers University
Contact:

Re: inert passive tracer boundary not being read

#8 Post by arango » Thu Nov 20, 2014 12:07 am

Please update your codes. I corrected this bug in the repository. See following :arrow: trac ticket for details.

cschultz
Posts: 13
Joined: Wed May 27, 2009 10:43 pm
Location: WHOI/MIT

Re: inert passive tracer boundary not being read

#9 Post by cschultz » Thu Nov 20, 2014 3:14 pm

Thank you for the help, really appreciate it!

Post Reply