What should be the initial conditions for SPIN-UP?

General scientific issues regarding ROMS

Moderators: arango, robertson

Post Reply
Message
Author
jayasankar
Posts: 15
Joined: Fri Sep 20, 2013 1:09 pm
Location: India

What should be the initial conditions for SPIN-UP?

#1 Post by jayasankar » Thu Sep 21, 2017 12:02 pm

HI all,

I have to do a spin-up, and I am trying with a 3 yr simulation (say 1/1990 to 12/1992) in loop such that 10 such loops would give me a 30 year Spin-up.

The boundary files are monthly data from a Global Ocean Model and surface fluxes are of 6-hrs. The global model is in 1 degree resolution and the regional model is 0.25 degree.

I am confused if to use climatology as initial conditions or conditions extracted from the global model.
From what ever I have read I think its better to use climatology with zero sea level anomaly and u and v velocities. If given climatology, the sea level difference at the boundaries would be high initially, I felt like may be sea level, temp and salt shall be extracted from global model but may be keep the u and v values to zero all over the basin. I would be grateful if anyone could give their view point regarding the same.


Thank you

User avatar
kate
Posts: 3803
Joined: Wed Jul 02, 2003 5:29 pm
Location: CFOS/UAF, USA

Re: What should be the initial conditions for SPIN-UP?

#2 Post by kate » Thu Sep 21, 2017 4:48 pm

I like to use consistent initial conditions, all from the same model, including velocities. But after a 30 year spin-up, it can't possibly matter that much (unless you do something foolish to make it blow up in the beginning).

jayasankar
Posts: 15
Joined: Fri Sep 20, 2013 1:09 pm
Location: India

Re: What should be the initial conditions for SPIN-UP?

#3 Post by jayasankar » Thu Sep 21, 2017 4:57 pm

Hi Kate,

Thanks for the reply

I have tried with initial conditions extracted from Global model. To my surprise ended up in BLOW-UP on 28th (9*3+1 th year) year. I suspect its the initial conditions.

Jayasankar

User avatar
kate
Posts: 3803
Joined: Wed Jul 02, 2003 5:29 pm
Location: CFOS/UAF, USA

Re: What should be the initial conditions for SPIN-UP?

#4 Post by kate » Thu Sep 21, 2017 5:04 pm

jayasankar wrote:I suspect its the initial conditions.
I very much doubt it, after all that run time. Can you get through the trouble spot with a shorter timestep? I often can. Yes, the model has random instabilities, at least in my experience. Did you investigate the nature of the blow-up? Mine are usually mid-depth at one isolated point. One wonders if the French/Sasha branches have solved this.

jayasankar
Posts: 15
Joined: Fri Sep 20, 2013 1:09 pm
Location: India

Re: What should be the initial conditions for SPIN-UP?

#5 Post by jayasankar » Thu Sep 21, 2017 5:09 pm

No, shorter time step is not helping. I was using 2400 sec (res- 0.2-0.25 deg) as time step and I even tried up to 600 sec

I didn't investigate the blow up properly, I have to do that.

jayasankar
Posts: 15
Joined: Fri Sep 20, 2013 1:09 pm
Location: India

Re: What should be the initial conditions for SPIN-UP?

#6 Post by jayasankar » Thu Sep 21, 2017 5:48 pm

hello Kate,

I am so sorry, I think I have wasted your time for no reason. The blow-up happened was because, the file got corrupted in transfer (there was a small miss match in size). I changed it and its working with out any issues. My apologies and million thanks for your help and suggestions on a number of occasions which were really helpful.

Jayasankar

Post Reply