-
Notifications
You must be signed in to change notification settings - Fork 543
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Init bugs2 #1142
base: develop
Are you sure you want to change the base?
Init bugs2 #1142
Conversation
@aronroland thank you for submitting this PR! As soon as the testing information from ERDC is added we'll start the tests and review. |
Converting this PR to a draft. Please mark as ready to review with testing information is available. @aronroland @thesser1 |
@aronroland @thesser1 will you help make sure that the description of this PR includes all the updates in this branch or confirm that it's all there already? |
sure ... |
Please see attached results from regtests. Looks like binary differences. |
Thanks for adding the testing @thesser1, I'll start working on it. |
@thesser1 is this ready to be marked ready for review? I did see that some tests had differences in the output of the unstructured grid cases, is this expected with the bug fixes? |
THR = DBLE(1.E-15) | ||
IF (SUM(A) .LT. THR) RETURN | ||
IF (EMEAN .LT. TINY(1.d0)) THEN | ||
S = 0 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
S
and D
are zeroed above. Why zero again here?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
yes, this was some leftover i cleaned it now, thanks @ukmo-ccbunney
@thesser1, I'm going to pause looking at this while there are some questions from Jessica and Chris. |
All, I had an issue with my netcdf library when I ran matrix.comp. I just noticed it and reran the comparison, and it increased the number of differences. Sorry for posting before checking. Based on comments above, it looks like there will be a fix and rerun needed, but I wanted to make sure the files posted here reflected all regression issues. I am sending differences to Aron, but I would keep this as draft for the time being. |
Thanks for the update, @thesser1. Please keep us posted |
Hi @JessicaMeixner-NOAA, we experience all HPCF maintenance. I guess it will be next week. Have a good weekend! |
@aronroland thanks for the update! |
@aronroland - I just wanted to check in to see if there were any updates on this PR. |
Pull Request Summary
Solves uninitialized variables issues and improves/fixes issues with the Neuman boundary condition.
Description
This pull request, solves uninitialized variable issues for TR0 and other bugs connected to initialization issues. Moreover, adding Neumann-type boundary conditions in the RD framework can be accomplished by calculating the solution at the boundary, without modifying the scheme itself. Specifically, a zero-gradient Neumann condition is applied at the boundary. However, this approach becomes less straightforward when dealing with positive source term contributions or phenomena like refraction and frequency shifting. In such scenarios, the Neumann boundary condition, characterized by a zero gradient, is not sufficiently defined. This is because the waves at the boundary may grow, change direction, or modulate, leading to a net influx of energy into the domain. To resolve this, we have developed a solution that involves bypassing the calculation of spectral propagation at Neumann boundary points and excluding the computation of positive source terms.
Please also include the following information:
The code changes are trivial.
It was a bug.
Answers may change for the regtest, where the certain settings are used.
Issue(s) addressed
none issue was raised yet.
Commit Message
Fixing uninitialized issues within the implicit scheme
Check list
Testing
I have checked on certain setups by SHOM (with Heloise Michaud), that this fixes the issues she has found.
Will be done by ERDC.
Mostly none ...
Will be done by ERDC