feat(MEGAN): Adding MEGAN and BDSNP biogenics and soilnox

[bbakernoaa]

This pull request introduces major enhancements to the CECE model's physics and chemistry capabilities, focusing on expanding and biogenic VOC speciation support. Key updates include the addition of biogenic emission modules, comprehensive speciation mapping files for several chemical mechanisms, and new unit/property tests for these features. There are also minor configuration and formatting improvements.

Major new features and modules:

• Introduced new biogenic emission and speciation modules: cece_megan3, cece_speciation_config, cece_speciation_engine, cece_canopy_model, cece_emission_activity, and cece_bdsnp, enhancing support for advanced biogenic VOC and soil NOx emissions. [1] [2]

Speciation mapping and mechanism support:

• Added comprehensive speciation mapping YAML files for MEGAN3-to-mechanism mappings: map_cb6.yaml, map_cracmm.yaml, map_racm2.yaml, and map_saprc07.yaml, supporting CB6_AE7, CRACMM2, RACM2, and SAPRC07 mechanisms, respectively. These files enable detailed and flexible mapping of emission classes to chemical mechanism species. [1] [2] [3] [4]
• Added a species definition file for the CB6_AE7 mechanism (spc_cb6.yaml), listing all mechanism species and their molecular weights for use in speciation calculations.

Testing and quality assurance:

• Introduced new unit/property tests for the added speciation configuration (test_speciation), MEGAN3 (test_megan3), and BDSNP soil NOx (test_bdsnp), ensuring correctness and robustness of the new modules. These tests are now included in the test discovery and build process. [1] [2]

[quaz115]

Thanks for putting this together. I can see the MEGAN3 canopy model with the gamma activity factors (LAI, age, soil moisture, temperature, PAR, CO₂), the BDSNP Fortran kernel with a C++ bridge, and the speciation engine framework intended to support CB6/CRACMM2/RACM2/SAPRC07 via the YAML mapping configs. Good call separating the speciation into configurable YAML files rather than hardcoding, that'll make adding new mechanisms straightforward. These are all great foundations. I also see the dust schemes (FENGSHA, Ginoux, K14) in the earlier commits.

I plan to start pushing code updates based on my recent BDISNP work. These should layer cleanly on top of your BDSNP MEGAN3 baseline (will think of a YAML toggle to switch between base and update if it makes computational sense).

Will start with detailed itemized review comments (and BDISNP code update) next week.

[bbakernoaa]

Thanks @quaz115. With this there is actually 2 yaml files 1) the MICM species yaml that includes the molecular weights and 2) the emission mapping per dataset/scheme. I’m still working out exactly how this would work on the anthropogenic side but the bones are there.

At the end of the call it will create output species like “MEGAN_ISOP” or “MEGAN_TERP” etc speciated from the 19 base megan species.

For the physics processes this is fairly clear cut in my opinion. But for anthro I’m still figuring it out

[bbakernoaa]

Note that this builds on #5 an #6 so it shows a ton of files changed but it's really only like 20 including the yamls and docs.

[bbakernoaa]

And the clang-format and clang-tidy calls modified a bunch too because I increased the line length to 150.