Problem of running ROMS with init and BC from a donor grid

[youngstor]

Dear ROMS forum,

I am trying to run a simulation for an area located on the US west coast. I extracted initial condition and boundary condition from a daily averaged solution of a donor grid (~3km resolution), and used it to force my 1km grid.

I ran the 1km grid for about 1 week, and had a look at the flow evolution (depth avg. mean current speed). Then I compared it with the 3km donor grid solution.

Here is the 3km donor grid flow evolution video (on Google Drive, will automatically play in a youtube window):
https://drive.google.com/file/d/0B4lTWc ... sp=sharing

Here is the evolution vidoe for the 1km grid that I ran:
https://drive.google.com/file/d/0B4lTWc ... sp=sharing

In the 3km grid, the flows comes in from south boundary and goes out from north boundary. However, in the 1km grid run, two strong eddies (one in the lower right area, and one in the left open area) occurred and the flow doesn't seem to penetrate the boundaries very well.

I played with different types of BC and nudging parameters, but wasn't able to fix it. I've also attached the *.h header file

altamaha.h

cpp header file

(2.34 KiB) Downloaded 268 times

with all the cpp, and the ocean.in

ocean.in

ocean.in input file

(64.58 KiB) Downloaded 238 times

file for your reference.

I appreciate your help and comments.

Thanks,
Xiufeng