Bugfix for dam break tests #451
Conversation
|
Note: The results of this test case are not BFB with master prior to E3SM-Project/E3SM#5154, but the general behavior with respect the the SSH field is unchanged. |
There was a problem hiding this comment.
Approved based on testing this on Compy in conjunction with E3SM-Project/E3SM#5279. Thanks @cbegeman. Sorry again for the delay on this.
There was a problem hiding this comment.
This looks great to me! I tested along with E3SM-Project/E3SM#5279 on Chrysalis with Intel and Intel-MPI. I tested both 40 and 120 cm cases.
I tested the validation as well (by rerunning the test and comparing with the previous run) and that also worked as expected.
|
I will merge this as soon as E3SM-Project/E3SM#5279 has been merged and we update the |
|
Submodule updated in #461 to include E3SM-Project/E3SM#5279 |
|
Thanks @cbegeman!!! |
In order to avoid crashes of the dam break test case, tracers need to be initialized (currently handled in init mode and fixed here E3SM-Ocean-Discussion/E3SM#33), and those tracers also need to be stored in the initial state stream (fixed here).
Related changes also included here are (1) adding the validation step to the test case and (2) disabling tracer tendencies, which are not needed in this test case.