Fix bug in aerinterp read routines that break 32-bit physics and pass NetCDF errors up through errflg&errmsg

[SamuelTrahanNOAA]

The aerinterp routines use real*4 arrays to read real*8 data when compiled with 32-bit physics. This causes a crash because of the NetCDF Fortran 77 interface, which does not know which datatype it has. These changes:

1. Switch to the Fortran 90 interface which can do type conversion when reading.
2. Detects NetCDF errors and returns them in errflg and errmsg.

• Fixes 32-bit physics crashes in aerinterp due to type mismatch in NetCDF calls. #83
• Fixes aerinterp routines do not pass NetCDF errors up through errmsg and errflg #84

[grantfirl]

Looks good to me.

[AnningCheng-NOAA]

looks good to me. Approved

[dustinswales]

@SamuelTrahanNOAA Thanks for these changes.

[SamuelTrahanNOAA]

@AnningCheng-NOAA

looks good to me. Approved

You didn't actually approve the PR. You just typed the word "Approved." To approve the pull request, you need to submit an approving review.

[AnningCheng-NOAA]

looks good. @SamuelTrahanNOAA, Thanks.

[grantfirl]

@SamuelTrahanNOAA Please merge in the latest ufs/dev branch in preparation for the merge queue.

[SamuelTrahanNOAA]

Please merge in the latest ufs/dev branch in preparation for the merge queue.

I already did that. Also, I validated against the baseline after merging ufs/dev.

[grantfirl]

Please merge in the latest ufs/dev branch in preparation for the merge queue.

I already did that. Also, I validated against the baseline after merging ufs/dev.

@SamuelTrahanNOAA Ah, I see that now. Sorry for spamming and thanks for being on top of it.

[zach1221]

Testing is now complete on UFS-WM PR #1797. @SamuelTrahanNOAA @grantfirl as this ccpp-physics sub-component PR is approved, please feel free to merge it.