Fix a logic for using specific humidity or mixing ratio directly

[weiwangncar]

TYPE: bug fix

KEYWORDS: real, use_sh_qv

SOURCE: internal

DESCRIPTION OF CHANGES:
The logic to control whether specific humidity or mixing ratio is used instead of RH is wrong (PR-1959). It is corrected in this PR.

LIST OF MODIFIED FILES:
M dyn_em/module_initialize_real.F

TESTS CONDUCTED:

1. Tested before and after this change. The data without SH is not affected, and the data with SH (EC model level data) is now used directly if use_sh_qv is set to true.
2. The Jenkins tests are all passing.

RELEASE NOTE:

[weiwangncar]

The regression test results:

Test Type              | Expected  | Received |  Failed
= = = = = = = = = = = = = = = = = = = = = = = =  = = = =
Number of Tests        : 23           24
Number of Builds       : 60           57
Number of Simulations  : 158           150        0
Number of Comparisons  : 95           86        0

Failed Simulations are: 
None
Which comparisons are not bit-for-bit: 
None

[dudhia]

So you don't need to test on those flags?

[weiwangncar]

So you don't need to test on those flags?

No.

[weiwangncar]

@kkeene44 @islas Can you review and approve this PR?

[kkeene44]

@weiwangncar
I'm just curious what happened before the fix. Would something fail, down the line?

[weiwangncar]

@weiwangncar I'm just curious what happened before the fix. Would something fail, down the line?

real.exe would not fail. When the input has one of the SH or Qv input, it doesn't matter if one sets use_sh_qv or not, it will always use computed RH (if RH isn't in the input. If RH is in the input, it uses it too).

[dudhia]

Does this fix anything in the last release or is it just fixing things added since?

[weiwangncar]

@dudhia This change is only added for this release.