update ERF interface for snow and wetland noahmp physics update

[cenlinhe]

This PR is to update the ERF/NoahMP driver interface to accommodate the recent NoahMP source code physics updates for snowpack and wetland processes.
The NoahMP source code updates can be found from these PRs:
#187
#186
#175
#174

[cenlinhe]

@akashdhruv Could you please do a ERF/NoahMP test for this updated version to ensure the updated driver interface works successfully?

[akashdhruv]

@cenlinhe Yes will do.

[cenlinhe]

akashdhruv

Hi @akashdhruv If you could test this at your earliest convenience, that would be great. We need to merge this PR asap to prepare for the Noah-MP release. Thanks!

[akashdhruv]

@cenlinhe I will finish this by Tuesday.

[akashdhruv]

@cenlinhe I was able to review today and here are my findings.

• Compilation error in src/SnowRadiationSnicarMod.F90 line 568. Changing to if (FlagSnicarUseAerosol .eqv. .true.) then fixed it.

• Compilation error in src/SnowRadiationSnicarMod.F90 line 571. Changing to if (FlagSnicarUseOC .eqv. .true.) then fixed it.

• Compilation error in src/SnowRadiationSnicarMod.F90 line 1522. Changing to write (*,"(a,e13.6,i0,i0,i0,i0)") "SNICAR ERROR: negative absoption : ", & fixed it.

• Segfault in src/SurfaceAlbedoGlacierMod.F90 line 59 and 68. Commenting them out fixed it.

• Segfault in src/SurfaceAlbedoMod.F90 line 109 and 110. Commenting them out fixed it.

[cenlinhe]

@akashdhruv Thank you for testing.

1. I assume that the compilation error is because you need to add the two new flag variables on your ERF model code side, right?
2. What are the segfault messages? Is it related to ERF specific setup or missing variables? These two new variables are related to our newly included snow physics in source code (src/SurfaceAlbedoGlacierMod.F90 and src/SurfaceAlbedoMod.F90), could you please find a fix without commenting them out? We hope the source code (src/) will be the same for all parent models using Noah-MP. In our offline Noah-MP test, these two places do not have issues and the model runs well.

[akashdhruv]

1. I assume that the compilation error is because you need to add the two new flag variables on your ERF model code side, right?

I don't think that is the case.

2. What are the segfault messages? Is it related to ERF specific setup?

The segfault creates a Backtrace which I was able to trace it back to these lines. I suspect that I am not supplying certain variables in my namelist file. Right now I set NSNOW=4 in my namelist file. Could that be causing an issue?

[akashdhruv]

If you can fix the compilation errors based on the suggestions that worked for me, then I think this is ok. I can review the segfaults carefully and fix them on erf side.

[cenlinhe]

Regarding the compilation, please ignore my message. I figured it out.
Regarding the segfault messages, yes, the NSNOW=4 might be the issue. In our Noah-MP model, currently we can only handle NSNOW = 3.

[cenlinhe]

If you can fix the compilation errors based on the suggestions that worked for me, then I think this is ok. I can review the segfaults carefully and fix them on erf side.

Yes, I will fix the compilation errors based on your suggestions. Thanks!

[akashdhruv]

I fixed the segfaults by changing:

SNOW_ALBEDO_OPTION = 2 --> SNOW_ALBEDO_OPTION = 3 in my ERF namelist file.

[cenlinhe]

@RAbolafiaRosenzweig @tslin2 I have fixed all the bugs reported by @akashdhruv above in the latest commit. Please review and approve this PR at your earliest convenience.
@akashdhruv I have fixed the bug that causes the segfault error. Now you should still be able to use SNOW_ALBEDO_OPTION = 2 in your erf namelist. The reason for the error is that those two new variables are only defined when SNOW_ALBEDO_OPTION = 3. Now I moved their initialization to within the SNOW_ALBEDO_OPTION = 3 if-statement.