Samuel Trahan (NOAA contractor)

> @SamuelTrahanNOAA Note that the main branch does not yet run in single precision either. This is the next major task but likely won't happen until August. I have updated...

The way I set it up, the stochastic physics uses the kind_phys from fv3, except for calculations that must be double precision. That's because I didn't realize the ocean called...

To clarify, stochastic_physics uses the same C preprocessor flag as FV3 (`#ifdef CCPP_32BIT`) to decide on single vs. double precision, regardless of what MOM6 wants.

@junwang-noaa This was NOT fixed in #1257. Please re-open this issue so I don't have to make a new one.

Actually, the hrrr_control variants already worked, they just weren't enabled. The reproducibility fix in that PR was for the rap_decomp.

No. This problem is not resolved.

I can fix the debug and 2threads variants in this PR: https://github.com/ufs-community/ufs-weather-model/pull/1437 Sadly, as yet, I have no fix for the restart or decomp variants. However, I suspect this bug...

When did this start happening?

I need to know the specific PR that broke it.

PR 2098 changes some NSSL microphysics code. The regression test never uses that code. It is likely that either: 1. A prior PR broke it, or 2. This problem has...