To use MEGAN 3.2 with CMAQ for a domain that is not 12US1, 108 HEMI, or the benchmark cases, you only need to run the MEGAN prep code, the EFP, and the code that converts csv files to IOAPI.
The rest of MEGAN 3.2 (MEGSEA, MEGCAN, MEGVEA, MGN2MECH, DAYMET) will be handled when you run CMAQ, including the time-dependent parts that need WRF/MCIP files.
I encourage anyone using MEGAN 3.2 to read the brief summary of its performance here:
The main points are that MEGAN 3.2 produces a lot of monoterpenes (strongly increases PM 2.5), and the MEGAN-YL scheme simulates less soil NO than BEIS-YL. MEGAN-BDSNP produces much more soil NO than either YL option (strongly affects ozone).
I will post the following here for prospective MEGAN 3.2 users. The MEGAN developers’ website currently states:
The code and input files are available for two versions of the MEGAN model. At the present time, most users should choose to use the MEGAN2.1 version.
When most people hear MEGAN they’re thinking of the popular MEGAN version 2.1.
We have made the code and input data for MEGAN3.2 available for anyone interested but you should note that there is no MEGAN3 publication or User Guide and we are not able to provide any MEGAN3 support for the general public at this time. We are working on finalizing MEGAN3 and expect to have a publication and User Guide available sometime in the near future.
In other words, this is an ongoing, cutting edge effort with publications and improvements imminent. Additional testing of MEGAN 3.2 is needed, welcome, and appreciated, but proceed with caution. MEGAN 3.2 in CMAQ is currently intended to serve modeling of domains outside of North America, sensitivity testing, and research specifically about biogenic emissions.