Fix bug in usage of FRSNO met-field

[lizziel]

Name and Institution (Required)

Name: Lizzie Lundgren
Institution: Harvard University

Describe the update

This PR fixes a bug where State_Met%FRSNO is inconsistently used. In some places it is used as fraction of grid box with snow covered land, while in other places it is used as fraction of land with snow cover. The GMAO meteorology field input to the model is defined as fraction of land with snow cover. To fix the issue we now define State_Met%FRSNO (renamed to State_Met%FRSNOW) as fraction grid box with snow cover on land, and compute it upon file read by multiplying the FRSNO import with fraction land import (FRLAND). Usage of the met-field throughout the model is now updated for consistency.

Also in this update is renaming State_Met%LANDIC to State_Met%LANDICE for clarity, and adding diagnostics for State_Met%IsWater, State_Met%IsLand, State_Met%IsIce, and State_Met%IsSnow for sanity checking. The surface type variables are type logical but the diagnostics are stored in State_Diag as REAL4.

Expected changes

This bug fix has the most impact for GCHP runs that use raw GMAO meteorology files. The raw files have missing values for all non-land grid boxes and FRSNO was set to 1e15 in those locations, resulting in incorrect logical masks for water, snow, and ice. The new multiplication by FRLAND removes that issue.

Mercury simulations are impacted because State_Met%FRSNO was used as if it were fraction grid box rather than fraction land.

Very small differences may be seen in GC-Classic full chemistry runs and GCHP runs that use processed meteorology files due to small changes in time-varying State_Met%FRSNOW when computing surface type masks (water, snow, ice, land). These changes would only be where snow is present.

Reference(s)

None

Related Github Issue(s)

• closes Wrong surface type field when using raw meteorology file GCHP#406

[yantosca]

I also approve! Thanks @lizziel.

[yantosca]

Integration tests are running

[yantosca]

All GEOS-Chem Classic execution tests passed:

==============================================================================
GEOS-Chem Classic: Execution Test Results

GCClassic #2fa1736 Update RTD docs for PR #2266 (Remove obsolete carbon sim options)
GEOS-Chem #203a480b6 Merge PR #2254 (Fix bug in usage of FRSNO met field)
HEMCO     #fddcae5 Update version to 3.8.1 for release

Using 24 OpenMP threads
Number of execution tests: 29

Submitted as SLURM job: 29651290
==============================================================================

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%  All execution tests passed!  %%%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

All GCHP integration tests passed:

==============================================================================
GCHP: Execution Test Results

GCHP      #e86ba15 GEOS-Chem submod update: Merge PR #2266 (Remove unused carbon sim options)
GEOS-Chem #203a480b6 Merge PR #2254 (Fix bug in usage of FRSNO met field)
HEMCO     #fddcae5 Update version to 3.8.1 for release

Number of execution tests: 11

Submitted as SLURM job: 29652416
==============================================================================

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%  All execution tests passed!  %%%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%