decreasing trend in bottom layer salinity ...

[turuncu]

Hi,

I have a weird results in my simulation. When i check the results of the model, the lowest layer salinity has decreasing trend in a specified region (see attached file, plot_dig_lev_1.pdf). The rest of the domain seems okay and also the main circulation patter seems correct. So, i don't have any idea about the reason. In this configuration, i am using 32 sigma layer with following options,

Vtransform == 2
Vstretching == 4
THETA_S == 5.0d0
THETA_B == 0.4d0
TCLINE == 10.0d0

and hc is 10 meter (the region average depth is 800 meter). I will try to increase the sigma layers and it might help but if you have any idea just let me know. By the way, i the bottom is parametrised as quadratic function and i set the RDRG2 as 0.02d0. Do you think that it causes the problem.

PS: The upper layers do not have same problem (see attachment, plot_dig_lev_11.pdf)

--ufuk

[kate]

Hi Ufuk,

I have no idea. You are somehow getting fresh water there. I doubt it has anything to do with the bottom drag. Can you look at vertical slices as a function of time and see where it's coming from? Is it stratified by temperature there or liable to mix vertically? You think it is being generated in situ?

[turuncu]

Hi Kate,

Actually this region is in the strait of sicily and it affects the stability of the model in decadal run. i think that it is not an observed process and just something goes wrong. The model circulation seems very wierd after 10-15 years of simulation. it accumulates and produces the problem. Now i am trying to increase number of vertical layer (40 layer) and the distribution of them. Do you think it helps? Then i will try it. i hope it works. i will keep posted about the problem. anyway thanks again.

--ufuk

[susonic]

Can I see your domain and the area problem occuring? Is it far from the boundary?

Have you ever changed the RDRG2 value? Mine is (0.0025 and the many test cases in ROMS are 0.003)

-JH

[turuncu]

Hi,

In my case, RDRG2 is little bit high (0.02d0). Now, i am trying to run the model with original value. I hope it works because i stuck into this problem. The region is the Sicily Strait in the Mediterranean Sea and it is not so close to the boundary. The z layer model (MITgcm) does not show this kind of trend in the same location. I'll keep you posted about the problem.

--ufuk