Constant value for all grid boxes in Passive Tracer monthly mean diagnostic

[lizziel]

Name and Institution (Required)

Name: Lizzie Lundgren
Institution: Harvard University

Description of your issue or question

The January monthly diagnostic of a MERRA2 benchmark after starting Jan 1 2019 contains all constant values of 1.00000640e-07 mol/mol up through the stratosphere. Running the simulation to get daily mean values for the same diagnostic shows heterogeneity with range 99.9990001 to 99.9990025 ppb in the zonal mean on various days, including day 1 and day 31. The daily means correspond well with restart files which have the same range for zonal mean.

Day 1 and Day 31 daily means, zonal mean:

Jan 1 and Feb 1 restart files, zonal mean:

I am creating this issue because it is strange the monthly diagnostic does not reflect what we see in the restart files and daily means. Is there a bug here? If it is a bug then it is not new to the latest version. Here is an example of the issue in the 14.2.0 benchmark. Notice that the GC-Classic diagnostic value matches what we see in the GCHP restart files and daily mean diagnostic. The GCHP constant value, however, is higher, and does not match the same pattern.

Full column zonal mean for Jan:

Surface mean for Jan:

[lizziel]

Note that this issue does not impact assessment of mass conservation in the transport tracer benchmark since we use restart files for the mass time series.