I am running CMAQv5.3.3 with ISAM module to trace source contribution to all PM2.5 species
The results in the basic CCTM_ACONC* and CCTM_CGRID* files looks correct. However, the results for AECI and AECJ of each source in CCTM_SA_ACONC* file are extremely high, reaching 1e15, since someday (2018194) in the simulation. Then we tried to set NEW_START to be TURE, the results became correct, so we think there is something wrong with initial condition.
And we find that since a few days before (2018189), the AECI and AECJ concentrations in the CCTM_SA_CGRID file at the top vertical layer has been extremely high, reaching 1e4. These high values appeared in the grids near domain boundary. We wonder how to fix this error.
These types of behaviors are difficult diagnose without trying to reproduce them using your data. In particular, AECI/J are typically the most behaved species because they pretty much just get emitted, blown around, and deposited. So I am not sure how to start except to ask if the base model concentrations in the CONC file appear to be correct.
Furthermore, have you tried the CMAQv5.4? We are recommending people upgrade to this version anyway, so is there a particular reason why are chose the older one?
I struggle with the same problem in CCTM_ACONC file with AECJ, ANAJ, and AORGCJ.
Have you found a solution by any chance?
I can’t figure out what is the problem.
I would appreciate a response.
(I have been using CMAQ version 5.4.)
Unfortunately, I could never reproduce the results reported by the initial poster, so I could not offer any solution. However, in you case, you mention that AORGCJ was problematic. While I am not sure about AECJ and ANAJ, the cloud SOA predictions did have an issue in the release version of the model for ISAM that was later addressed in 5.4+ branch. 5.4+ branch offers other minor bug fixes to the model in general and I recommend you try that version to see if you problems will be resolved.
I met the same problem with extreme high values of AORGCJ. In the ACONC files its value is normal, so I guess there will be an issue in the ISAM code, fyi.