VSC value needs updating for SMC regtests

[ukmo-ccbunney]

A recent update to the SMC grid to allow for wetting an drying (#1192) changed the behaviour of the cell depths with respect to the VSC depth scaling value.

This is impacting the results of one of the regression tests (ww3_tp2.10) as described here:
#1230 (comment)

Excerpt below:

This recent SMC grid update has activated an old conversion factor VSC to reverse the depth values stored in the SMC grid cell array before they are assigned to the WW3 ZB variable:

!/SMC !Li Minimum DMIN depth is used as well for SMC.
!/SMC !! ZB(ISEA)= - MAX( DMIN, FLOAT( IJKCel(5, ISEA) ) )
!/SMC !Li Allow land cell to be defined by ZLIM value and only reset
!/SMC !Li MAPST* land values for sea points. JGLi03Nov2023
!/SMC ZB(ISEA)= DVSMC*FLOAT( IJKCel(5, ISEA) )

The DVSMC is a copy of the VSC conversion factor, which is specified in the w3_grid.inp file on the old depth input file line:
-0.1 10.0 30 -1. 1 1 '(....)' 'NAME' 'S36125Depth.dat'
Where the fourth argument -1.(0) Is for the VSC. This factor is not defined in the namelist input file and it results in all wet points are converted into land points. The namelist input file in tp2.10 test needs to be updated for this factor. If the old ww3_grid.inp file is used, it should be fine.

From: Mickael Accensi @.>
Sent: Friday, May 31, 2024 8:17 AM
To: NOAA-EMC/WW3 @.>
Cc: JianGuo Li @.>; Mention @.>
Subject: Re: [NOAA-EMC/WW3] Add option to use NetCDF output instead of binary for point output (PR #1230)

matrix is done but I have both an error in develop and bin2nc branches for ww3_tp2.10 for output point processing due to the fact that the only requested output point (B45005) is considered on land. It is a SMC grid test so it is not linked to this PR. @ukmo-ccbunneyhttps://github.com/ukmo-ccbunney and @ukmo-jianguo-lihttps://github.com/ukmo-jianguo-li can probably double check this problem.

[ukmo-ccbunney]

This affects both ww3_tp2.10 and ww3_tp2.16 , but only when using the NML input file. The default value for DEPTH%SF=1.0 but the input files for those regtests require DEPTH%SF=-1.0

Running with INP files is not affected as the SF parameter is mandatory.

Fix coming imminently.