Tracers don't radiate out at boundary

[Lxc]

Dear all,
I'm running a climtology case by setting the atmos forcing and boundary with COADS and climtology HYCOM data respectively, the initial condition is also extracted from HYCOM. Tidal forcing is also included. After a few months, some strange features appears:
1. The flow encounters the eastern boundary (in Japan Sea) but don't radiate out, it flows along the open boundary and deflect back along the coast.
2. At Changjiang River mouth, the SST is weirdly low. I have turned off the temp river source, and I can't figure out why.

Can you help me?
Thanks in advance!


Here is my open boundary condition:
LBC(isFsur) == Clo Cha Cha Clo ! free-surface
LBC(isUbar) == Clo Fla Fla Clo ! 2D U-momentum
LBC(isVbar) == Clo Fla Fla Clo ! 2D V-momentum
LBC(isUvel) == Clo RadNud RadNud Clo ! 3D U-momentum
LBC(isVvel) == Clo RadNud RadNud Clo ! 3D V-momentum
LBC(isMtke) == Clo Clo Clo Clo ! mixing TKE

LBC(isTvar) == Clo RadNud RadNud Clo \ ! temperature
Clo RadNud RadNud Clo ! salinity

[kate]

Turning off the temperature of the river source is numerically unstable.

Do you mean that you use monthly means from HYCOM for the boundary files? I have always had better luck with more frequent boundary information and even then it's not perfect.

[Lxc]

Thank you for your reply Kate. Yes, I'm using the monthly HYCOM data as my boundary. I'will turn on the temperature river source and try different boundry.
And may I ask another question? I have set my Nudging/relaxation time scales all to 360 days, with the OBCFAC being 120. Could this be the problem?

[kate]

No, that's not the problem. I use those numbers too. I haven't tried monthly HYCOM, but back in the day we would try monthly Levitus and have a terrible time with the boundaries. I use daily HYCOM and I also nudge to monthly HYCOM in a band around the edge of the domain.

[Lxc]

Thank you Kate! I will try that.