prevent negative values for passive tracer concentration

asked 2021-10-13 18:38:23 +0200

Hi, I am running a simulation where I have a passive tracer with a non-zero concentration in a small portion of the domain and a zero concentration everywhere else. I would like to prevent negative values for the passive tracer concentration. I have tested different advection scheme (UP, WENO) and different mixing scheme (LMD, GLS) but they all yield pretty strong negative values. Is there a way to prevent those? Thank you, Bertrand

edit retag flag offensive close merge delete

add a comment

0

answered 2021-10-19 21:10:23 +0200

Bertrand Delorme
1

Interesting, let me look into that. Thank you, Xupeng.

@patrick: maybe something you would like to check as well?

edit flag offensive delete link

Comments

Hi Bertrand, I am curious to know if your issue has been resolved after changing the source code as described in my previous post?

xupeng66 ( 2021-10-21 17:41:08 +0200 )edit

add a comment

0

answered 2021-10-19 20:10:42 +0200

xupeng66
3

Bertrand,

I posted this: https://forum.croco-ocean.org/questio... a while ago about the issue I had with adding a passive tracer in my buoyant plume simulations. I later found that the source code that computes tracer surface flux needed to be modified. I made some changes in the code and the results looked reasonable afterwards. Maybe you are having the same issue?

edit flag offensive delete link

add a comment

0

answered 2021-10-19 18:31:58 +0200

Bertrand Delorme
1

Hi Patrick and Xupeng. I did resolve the confict so I am using only TS_HADV_WENO5, TS_VADV_WENO5, UV_HADV_WENO5, UV_VADV_WENO5 and BIO_HADV_WENO5 but negative values for the passive tracer concentration persists. Please let me know if you have any insights on how to avoid that. Thank you.

edit flag offensive delete link

add a comment

0

answered 2021-10-19 18:27:02 +0200

xupeng66
3

Hi Bertrand. I also have the same problem. Have you tried the 'BIO_HADV_WENO5' CPP option? The CROCO user guide says that advection scheme is used for passive tracer.

edit flag offensive delete link

add a comment

0

answered 2021-10-13 19:35:30 +0200

pmarches

66 http://www.legos.obs-m...

yes there is conflict of course (SPLINES will be chosen).

edit flag offensive delete link

add a comment

0

answered 2021-10-13 18:49:08 +0200

Bertrand Delorme
1

Hi Patrick. Yes, they are all defined. Note that I have both UV_VADV_SPLINES
and UV_VADV_WENO5 in the active CPP options. I wonder if there is a conflict there.