In order to track PM2.5 pollution sources, which species should be extracted from SA files

I used the chemical mechanism of CB06ea6 to run CMAQ-ISAM. I found that there was a large gap between the sum of ISAM labeled PM.25 species and the PM2.5 concentration extracted from conc files. Here are the extracted species from my conc file:
But in the SA file, my script tracks only the following species:
I don’t know if I’m tracking enough species.
And I don’t know if the SO2, SULF, SULRXN in SULFATE, and HNO3, NO, NO2, NO3, HONO, N2O5, PNA, PAN, PANX, NTR1, NTR2, INTR, CLNO2, CLNO3 in NITRATE, and NH3 in AMMONIUM, and HCLA25 in CHLORINE shoud be tracked when I trace the source of PM2.5?
To sum up, I don’t know whether it is caused by too few PM2.5 species tracked or species errors when extracting PM2.5 concentration in conc documents.
Thank you for your reply

Based on the mechanism you quote, it seems you may not be running the latest CMAQ version 5.4. In previous CMAQ version, ISAM was not available for secondary organic aerosols (SOA) which likely explains the mismatch between CONC and SA species you are reporting. ISAM for SOA was introduced in CMAQv5.4

In older version of CMAQ, the only available tag classes for PM were SULFATE, NITRATE, AMMONIUM, EC, OC, and PM25_IONS, with OC representing only primary organic aerosols.

In CMAQv5.4 supporting SOA, the OC tag class representing primary OC was replaced by the NVPOA tag class, and the additional tag classes OA_TOT and PM_TOT were added to also allow tracking of SOA species.

thanks you. But in the region I simulated, soa generation was not significant, and it was in the northwest of China,in winter. At the mean time, there is not enough time for me to switch to CMAQ5.4 again. I wonder if the PM2.5 species I tracked to SA files are enough?

If you consider SOA formation to be of low importance for your application, the subset of the CONC file PM species that are included in your CMAQv53 SA output file when enabling the SULFATE, NITRATE, AMMONIUM, EC, OC, and PM25_IONS tag classes should be sufficient. The remaining CONC file PM species (AALK1J, AALK2J, AXYL1J, AXYL2J, AXYL3J, ATOL1J, ATOL2J, ATOL3J, ABNZ1J, ABNZ2J, ABNZ3J, ATRP1J, ATRP2J, AISO1J, AISO2J, ASQTJ, AORGCJ, AISO3J, AOLGAJ, AOLGBJ) are SOA species, so if you do not consider them important for your application, you likely need not worry about their absence in your SA files.

Ok, I see. I’ll check my VOC emissions again. You answered a lot of my questions. Thank you very much.