ROMS<kshedstrom version> question about the "zeta <free surface elevation>"

[yangwu1215]

I'm recently studying the kshedstrom version <https://github.com/kshedstrom/roms/tree/master>, and trying to apply it on a freshwater lake.

I plot the evolutions of the 5-day averaged depth-averaged currents <ubar, vbar, vector>, free surface elevation <zeta, contour>, and water temperature along a southwest-northeast transection <indicated by the dashed line in the right panel>.

However, the spatiotemperal variations of the zeta seem quite weird during the whole simulation periods <20130501-20131231>, and I haven't figure it out.

Following the the animation.

Hope to get some help with this problem. Great thanks!

[kate]

Are the red areas elevated land, as in for wetting and drying? I too would have expected the currents to be more closely related to the ssh contours - it looks like they align right at the end?

[yangwu1215]

The red and wet contours in the right panel both denote the free-surface variations <zeta> in the whole simulation area.

The following is my model domain and Lake Nam Co depth.

model domain and lake depth.png (36.14 KiB) Viewed 2797 times

In winter, the currents mainly flow in response to the strong surface winds, which mainly directs eastward during winter.

I also utilized POM <Princeton Ocean Model> to simulate the lake thermohydrodynamics. The depth-averaged currents seem organized and match well with the free-surface elevation gradients. For now, I can't understand why the free-surface elevation modeled by ROMS is much bigger, almost 10 times larger than that in POM, and seems weird with little variations.

The following is the POM simulated animation of the 5-day averaged depth-averaged currents <ubar, vbar, vector>, free surface elevation <zeta, contour>, and water temperature along a southwest-northeast transection <indicated by the dashed line in the right panel>.

[kate]

What is it that you hope to gain by switching from POM to ROMS? I haven't modeled such a small lake with ROMS, but I expect it can handle it. It smells like a setup problem or who knows what. Anyway, I am in the process of leaving ROMS for yet another model - MOM6. It too should be able to handle a small lake.

[yangwu1215]

Compared with POM, ROMS is more applicable for coupling with the other models <WRF, CICE, SWAN>, which is more suitbale for my future research on the lake-air interactions.
For now, both POM and ROMS can well capture the lake temperature, while their performances in simulating the free-surface elevation and currents differ greatly. I want to seek for advice to handle the problem why the ROMS simulated free-surface elevation seems unreasonable.