Standard usage of the RHmin parameter

[ChristopherHill-NOAA]

The RHmin parameter exists within the UPP suite - under ncep_post.fd/params.F - as a standardized minimum possible value of relative humidity (RH), available for use throughout the code to prevent illegal mathematical calculation of a formula that could otherwise involve a zero- or negative-value representation of RH. The RHmin parameter predates the UPP suite and derives from the development of the CALRH subroutine within legacy post-processing code. The RHmin parameter is known to be used in some of the more commonly called routines under ncep_post.fd: FDLVL.f, MDL2P.f, MDL2THANDPV.f, MISCLN.f, and UPP_PHYSICS.f (which now stores CALRH and system-specific derivations).

The generation of this discussion was instigated by a review of the RHmin parameter value setting through PRs #1181 and #1186. As the GFS model now features a top vertical level of 80 km (mesopause), increased analytical scrutiny of formulaic calculations representing stratospheric physics requires a revision of the RHmin value setting from 1E-6 to a more representative value of 1E-7, as provided in #1181. The abovelisted routines referencing RHmin will make use of the new value.

Elsewhere in the UPP suite, there exist other routines that make use of similar logic to compare RH values to a minimum value, but do not use RHmin itself. Just as analysis of model output through UPP led to the RHmin value revision, so could additional analysis eventually warrant a standardization of the use of the RHmin parameter (and perhaps other parameters) in more routines. UPP routines employing alternate representations of a minimum RH value (as 0.01, except where noted) are provided below, and are intended for reference in any future issue or PR:

MDL2SIGMA.f : lines 233-236, 265-268, 518-521, 603-606, and 705

MDL2SIGMA2.f: lines 234-237

LFMFLD.f: lines 126-129

NGMFLD.f: lines 164-167

GFIP3.f line 123 (value 1E-6 is hard-coded here)