NaN or Infinity detected in VDIFF

CMAQ Run Time Errors and Issues
1

Hi, I am using CMAQv5.3.3 with enabling ISAM and my simulation fails on the 21th time step. The errors are as follows:
Processing Day/Time [YYYYDDD:HHMMSS]: 2020226:210000
Which is Equivalent to (UTC): 21:00:00 Thursday, Aug. 13, 2020
Time-Step Length (HHMMSS): 000640

 After        VDIFF :  Gas  7.862E+31  | Aer        NaN  | Non  1.014E+31
 
 *** ERROR ABORT in subroutine CKSUMMER
 NaN or Infinity detected in VDIFF
 Date and time  21:00:00  Aug. 13, 2020   (2020226:210000)
 Processing Day/Time [YYYYDDD:HHMMSS]: 2020226:210000
   Which is Equivalent to (UTC): 21:00:00 Thursday,  Aug. 13, 2020
   Time-Step Length (HHMMSS): 000640
 
 After        VDIFF :  Gas  7.862E+31  | Aer        NaN  | Non  1.014E+31
 
 *** ERROR ABORT in subroutine CKSUMMER
 NaN or Infinity detected in VDIFF
 Date and time  21:00:00  Aug. 13, 2020   (2020226:210000)

Any recommendations for how to solve this or investigate further?
Thanks!
CTM_LOG_000.v533_intel_Bench_2011_12SE1_20200813.txt (470.4 KB)

2

You might try a run that is compiled using the debug/check-everything compile-flags used in the corresponding I/O API *Makeinclude.dbg – that should give you the subroutine and line-number at which the NaN occurs.

3

Thanks. I figured out the issue.

4

could you please tell me what was the issue, cause im having the same problem!

5

@jstergiou please post your log files so that we can provide you with assistance.

6

Its ok i set CTM_ABFLUX to N and it worked! Thank you.

7

NaN can be tested by comparing the value against itself (they are defined to not be equal any values).

8

I figured out the issue...

9

Hi, I met the same issue with CMAQv5.4, and setting CTM_ABFLUX to N wouldn’t work for me. Here is my log file, any help would be grateful.
CTM_LOG_000.v54_intel_china27km_20220726.txt (130.7 KB)

10

I would recommend compiling CCTM in debug mode to find out where in the code the computations causing the NaN value occur - the same recommendation given by @cjcoats in the second post from the top.

11

Thank you for your advice, I will try.