Feature 1809 gen prob

[JohnHalleyGotway]

Pull Request Testing

@georgemccabe we'll likely need updates the tc-gen METplus wrapper to supply the new -edeck command line option and configure the new prob_genesis_thresh option. I'd recommend holding off until #1810 is done, which will also change the inputs/configuration options.

@halperin-erau please review the config file changes and doc updates, if possible:
https://met.readthedocs.io/en/feature_1809_gen_prob/Users_Guide/tc-gen.html
You can try out the actual code in the docker container for the MET-10.1.0-beta4 development release, once that's ready.

• Describe testing already performed for these changes:
  
  Manual testing of new functionality using sample files provided by @halperin-erau. During the beta5 development cycle, @KathrynNewman will be adding a METplus use case demonstrating this new functionality with a much larger sample of data. Recommend revisiting the existing tests in unit_tc_gen.xml and consider making them more robust once that larger dataset if available.

• Recommend testing for the reviewer(s) to perform, including the location of input datasets, and any additional instructions:
  
  Review the regression test output in kiowa:/d1/projects/MET/MET_pull_requests/met-10.1.0/met-10.1.0_beta4/feature_1809

• Do these changes include sufficient documentation updates, ensuring that no errors or warnings exist in the build of the documentation? [Yes]

Still working on the doc updates. Once they're complete I'll change this from a draft PR to one that's ready for review.
See doc updates here:
https://met.readthedocs.io/en/feature_1809_gen_prob/Users_Guide/tc-gen.html

• Do these changes include sufficient testing updates? [Yes]

Add a new call to tc_gen in unit_tc_gen.xml.

• Will this PR result in changes to the test suite? [Yes]
  
  If yes, describe the new output and/or changes to the existing output:
  

Expected changes are:
In kiowa:/d1/projects/MET/MET_pull_requests/met-10.1.0/met-10.1.0_beta4/feature_1809/test_regression.log

• new tc_gen probabilistic output files

ERROR: folder MET-develop/test_output missing 6 files
    tc_gen/tc_gen_prob_genmpr.txt 
    tc_gen/tc_gen_prob_pct.txt 
    tc_gen/tc_gen_prob_pjc.txt 
    tc_gen/tc_gen_prob_prc.txt 
    tc_gen/tc_gen_prob_pstd.txt 
    tc_gen/tc_gen_prob.stat 

• modification to the contents of existing GENMPR line type

file1: MET-feature_1809_gen_prob/test_output/tc_gen/tc_gen_2016_genmpr.txt
ERROR: compareStat() failed
--
file1: MET-feature_1809_gen_prob/test_output/tc_gen/tc_gen_2016.stat
ERROR: compareStat() failed

• Please complete this pull request review by [Monday 11/15/21].
  

Pull Request Checklist

See the METplus Workflow for details.

• Complete the PR definition above.
• Ensure the PR title matches the feature or bugfix branch name.
• Define the PR metadata, as permissions allow.
  Select: Reviewer(s)
  Select: Organization level software support Project or Repository level development cycle Project
  Select: Milestone as the version that will include these changes
• After submitting the PR, select Linked issues with the original issue number.
• After the PR is approved, merge your changes. If permissions do not allow this, request that the reviewer do the merge.
• Close the linked issue and delete your feature or bugfix branch from GitHub.

[KathrynNewman]

Quick question on consistency before I approve: I see PROB_GENESIS for the fcst_var and obs_var in the output files, which is consistent with the genesis methodology listed in the UG (GENESIS_DEV, GENESIS_OPS, or GENESIS_PROB). But in the TCGenConfig_prob the dev and ops scoring method flags appear (which I believe is what we want). How are these differentiated?

[JohnHalleyGotway]

@KathrynNewman you are correct that these flags are listed in TCGenConfig_prob:

dev_method_flag = TRUE;
ops_method_flag = TRUE;

However, these currently have no effect on the probability of genesis verification! These is only one set of vx logic defined for this data, as described in the Scoring Logic section.

So we do NOT have different "development" and "operational" scoring logic. That was a detail that was a bit murky for me. Should we support 2 types of scoring logic for this input? If so, is it crystal clear to you what they should be and how they should differ?

[KathrynNewman]

I'm sorry for the stupid question, John! I was recalling that you listed both DEV and OPS in the meeting notes from 11/11, but now that I look back at the issue, I see I somehow missed the entire discussion between you and Dan on only using OPS=PROBGENESIS. This all sounds good!

[georgemccabe]

I created 2 METplus issues that are related to this work:
Enhance TCGen wrapper to add support for new configurations: dtcenter/METplus#1273
New Use Case: TCGen Verify Genesis Probabilities from ATCF e-deck files: dtcenter/METplus#1274