Fixing African easterly wave density plots in TC analysis

E3SM-Project / e3sm_diags

E3SM Diagnostics package

https://e3sm-project.github.io/e3sm_diags

BSD 3-Clause "New" or "Revised" License

42 stars 32 forks source link

Fixing African easterly wave density plots in TC analysis #851

Closed chengzhuzhang closed 1 month ago

chengzhuzhang commented 1 month ago

Description

Karthik (@karthik-balaguru) noted the odd Y shapes in easterly wave density plots in recent TC analysis results (issue 1). And @paullric also noticed that the ERA5 results are also off, which always have a base value of ~5 (issue 2).

Closes #

Checklist

[ ] My code follows the style guidelines of this project
[ ] I have performed a self-review of my own code
[ ] My changes generate no new warnings
[ ] Any dependent changes have been merged and published in downstream modules

If applicable:

[ ] New and existing unit tests pass with my changes (locally and CI/CD build)
[ ] I have added tests that prove my fix is effective or that my feature works
[ ] I have commented my code, particularly in hard-to-understand areas
[ ] I have made corresponding changes to the documentation
[ ] I have noted that this is a breaking change for a major release (fix or feature that would cause existing functionality to not work as expected)

chengzhuzhang commented 1 month ago

Not sure the root cause of issue 2, but I think plotting libraries changes caused the second issue over versions. By reading in files without subsetting the region, it fixed the issue. (Now, the region is bounded only when plotting)

Issue 1 remains, I will keep investigating.

chengzhuzhang commented 1 month ago

Comparing with apply tempest-extremes (TE) on native ne120pg2 vs data regrided to 720x1440. The Y shapes only appear with native grid data. Below are the AEW density plots. awe_ne120pg2 awe_720x1440

The AEW density plots uses three TE functions: VariableProcessor to calculate relative vorticity DetectNodes and StitchNodes to detect and stitch candidates Finally HistogramNodes to generate histogram for the detection. Detailed steps can be found here

@paullric I'm not sure if TE is applied properly in this case, any advice for further trouble shooting is appreciated!

chengzhuzhang commented 1 month ago

Also to add that: TC count increases if the storms are tracked on the regridded output. ~73 (regridded data on 720x1440) vs ~67 (native ne120) for one year of data.

chengzhuzhang commented 1 month ago

Also to add that: TC count increases if the storms are tracked on the regridded output. ~73 (regridded data on 720x1440) vs ~67 (native ne120) for one year of data.

Following suggestion from @wlin7, when regridding, with a bilinear mapping file (to replace the aave one I used), the TC count (70) get closer to the results tracked with native ne120 data.

chengzhuzhang commented 1 month ago

This PR will also address: empty input TC files. from comment: https://github.com/E3SM-Project/zppy/pull/633#issuecomment-2423038746

chengzhuzhang commented 1 month ago

This PR will also address: empty input TC files. from comment: E3SM-Project/zppy#633 (comment)

This is addressed by https://github.com/E3SM-Project/e3sm_diags/pull/866/commits/0cb36e971c32db4caa3002bc184fff2e1df895db by @tomvothecoder in cdat-migration-branch. I will copy the code over here, so that current main branch also has this fix.

chengzhuzhang commented 1 month ago

@tomvothecoder this PR is ready to review, it partially addresses the plotting problem (Y shapes remains); also cross year TCs are removed; I copied over the error capture for cyclones stitches files as well.