Skip to content
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

Non orographic gw parameterization for GCM configurations #983

Merged
merged 1 commit into from
Oct 28, 2022

Conversation

jiahe23
Copy link
Contributor

@jiahe23 jiahe23 commented Oct 19, 2022

Purpose

This PR integrates GW parameterization with GCM configuration, and adds an example into buildkite.

Differences in its integration into single column (as described in the paper) vs gcm (as in GFDL codes) configurations:

  • source level identification: fixed height in single column vs fixed pressure in gcm;
  • source amplitude: fixed number in single column vs latitude dependency in gcm.

These differences require different cache variables and forcing computations for the two setups.

Closes #783

@jiahe23
Copy link
Contributor Author

jiahe23 commented Oct 19, 2022

@charleskawczynski I know I allocate a lot here... Could you help take a look together to improve the performance?

@charleskawczynski
Copy link
Member

What's the difference between the sphere and column gravity wave parameterizations?

@charleskawczynski
Copy link
Member

Would it be easier to revisit this after revisiting the existing GWP first?

Copy link
Contributor

@LenkaNovak LenkaNovak left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This looks great, @jiahe23 ! When comparing the sponge vs GW cases in Buildkite, I guess the difference is what we would expect? A slightly less damped but stable run. So without the GW (or sponge) the same setup blows up?

@jiahe23
Copy link
Contributor Author

jiahe23 commented Oct 20, 2022

This looks great, @jiahe23 ! When comparing the sponge vs GW cases in Buildkite, I guess the difference is what we would expect? A slightly less damped but stable run. So without the GW (or sponge) the same setup blows up?

The two links do not seem to work ... I added it into the short ci runs to make sure the functionality works. We will need to add a longrun experiment to see if it stables the longrun (hundreds of days at least)

@LenkaNovak
Copy link
Contributor

LenkaNovak commented Oct 20, 2022

This looks great, @jiahe23 ! When comparing the sponge vs GW cases in Buildkite, I guess the difference is what we would expect? A slightly less damped but stable run. So without the GW (or sponge) the same setup blows up?

The two links do not seem to work ... I added it into the short ci runs to make sure the functionality works. We will need to add a longrun experiment to see if it stables the longrun (hundreds of days at least)

Ahh, I was just linking the u plots of the sponge and GW runs on Buildkite. So if we don't have a physical validation case, it would be better to compare this run with a control run (i.e., hightop with no sponge). Otherwise it's difficult to see what difference it makes (unless the control is hidden somewhere in the Buildkite list and I'm not seeing it...). :)

@jiahe23 jiahe23 force-pushed the jh-zs-ln/GW-lat branch 2 times, most recently from b1d18a1 to e33b73a Compare October 27, 2022 17:03
@jiahe23 jiahe23 force-pushed the jh-zs-ln/GW-lat branch 3 times, most recently from 0ac28fe to ef9e5d7 Compare October 27, 2022 19:18
@jiahe23
Copy link
Contributor Author

jiahe23 commented Oct 27, 2022

bors try

bors bot added a commit that referenced this pull request Oct 27, 2022
@jiahe23
Copy link
Contributor Author

jiahe23 commented Oct 27, 2022

bors try-

@jiahe23
Copy link
Contributor Author

jiahe23 commented Oct 27, 2022

bors try

bors bot added a commit that referenced this pull request Oct 27, 2022
@jiahe23
Copy link
Contributor Author

jiahe23 commented Oct 27, 2022

bors try-

@jiahe23
Copy link
Contributor Author

jiahe23 commented Oct 27, 2022

bors try

bors bot added a commit that referenced this pull request Oct 27, 2022
@bors
Copy link
Contributor

bors bot commented Oct 27, 2022

try

Build succeeded:

@jiahe23
Copy link
Contributor Author

jiahe23 commented Oct 27, 2022

bors r+

bors bot added a commit that referenced this pull request Oct 27, 2022
983: Non orographic gw parameterization for GCM configurations r=jiahe23 a=jiahe23

## Purpose 
This PR integrates GW parameterization with GCM configuration, and adds an example into buildkite.

Differences in its integration into single column (as described in the paper) vs gcm (as in GFDL codes) configurations:
- source level identification: fixed height in single column vs fixed pressure in gcm;
- source amplitude: fixed number in single column vs latitude dependency in gcm.

These differences require different cache variables and forcing computations for the two setups.


Closes #783 


Co-authored-by: Jia He <[email protected]>
@jiahe23
Copy link
Contributor Author

jiahe23 commented Oct 27, 2022

bors r-

@bors
Copy link
Contributor

bors bot commented Oct 27, 2022

Canceled.

@jiahe23
Copy link
Contributor Author

jiahe23 commented Oct 27, 2022

bors r+

@bors
Copy link
Contributor

bors bot commented Oct 28, 2022

Build succeeded:

@bors bors bot merged commit 1cedb39 into main Oct 28, 2022
@bors bors bot deleted the jh-zs-ln/GW-lat branch October 28, 2022 00:20
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Implement the non-orographic gravity wave parameterization
4 participants