DDMv5.4 variable names from sensitivities in CTM_SDRYDEP_1 file


I am running DDM5.4 and looking at the CTM LOG files of the first day (which ran to completion). I noticed that CMAQ is trying to write out a variable name, which were not specified anywhere in the run script or in the dat file to the output file CCTM_SENDDEP (i.e. referred to as variable CTM_SDRYDEP_1, in the log script)

For instance, in a case of four sensitivities labeled as ‘COA, MET, FOI OTH’, CMAQ appears to be trying to write a variable of name NO2_COA_MET_FOI_OTH** to the file CTM_SDRYDEP_1

It appears that this error is only in the first LOG file, CTM_LOG_000 file, not the others.

Hi Abi,

Thank you for noticing and reporting this issue. There is, indeed, a bug in writing out these deposition sensitivities. We do have a fix that will be released at some point soon in the 5.4+ version.

Unfortunately, this bug makes deposition sensitivity output unusable, so if you’d like to analyze these before the fix is released, you need to go into the file vdiffproc.F and make 2 small changes.

1st, on line 761, replace variable SPCNAME with SPCBASE:
SPCBASE = DDEP_SPC( V )( 1:12 )
2nd, on line 763, replace variable SPCNAME with SPCBASE in the 2nd occurrence:

Then, remove the old object files (make clean) and recompile and it should generate correct deposition sensitivity output data.


Thank you so much for this Sergey.

I gave it a try but I ran into another snag.

This is one of the errors “This name does not have a type and must have an explicit type. [SPCBASE]”

I don’t think I need to worry about the dep sensitives at the moment, so not so critical. But it would be nice to stop the error if possible.

Yes - I am sorry. The variable also needs to be defined in that same subroutine. At the top, at the end of “Local Variables” please add:



No problem. My Fortran skills could do with some brushing up (basically non-existent at this point), so I’m quite grateful for your tips!!