sss addesd to sfc_data bkg files

[HamidehGMAO]

Description

I have added sss to the gfs_c12 bkg data for the 2020121500. Previously I had added sss using the World Ocean Atlas (WOA) data "https://www.ncei.noaa.gov/products/world-ocean-atlas", for the 20180415. As there is not any data available for after 2018 in WOA website, I have used the sss data from SMAP sss retrievals here "https://podaac.jpl.nasa.gov/dataset/SMAP_JPL_L3_SSS_CAP_MONTHLY_V5?ids=&values=&search=SMAP%20JPL%20V5%20L3&provider=PODAAC"

Issue(s) addressed

Add SSS field to background files #417

• fixes https://github.com/JCSDA-internal/fv3-jedi/issues/417

[cmgas]

Do you have a corresponding PR in fv3-jedi to add sss to the tests?

[HamidehGMAO]

Do you have a corresponding PR in fv3-jedi to add sss to the tests?

@cmgas I just made this PR https://github.com/JCSDA-internal/fv3-jedi/pull/503, but I couldn't find the test associated to the hofx_fv3-gfs. when I run the ctest in fv3-jedi just tier1 tests run, should I do something to run tier2?

[cmgas]

Do you have a corresponding PR in fv3-jedi to add sss to the tests?

@cmgas I just made this PR JCSDA-internal/fv3-jedi#503, but I couldn't find the test associated to the hofx_fv3-gfs. when I run the ctest in fv3-jedi just tier1 tests run, should I do something to run tier2?

You need to build your fv3-jedi repo with tier2 tests enabled. I can do that for you today!

[HamidehGMAO]

@cmgas Thanks. If you can point me to the instruction, or tell me the steps I will do that. Thanks

[cmgas]

@cmgas Thanks. If you can point me to the instruction, or tell me the steps I will do that. Thanks

You need to export the environment variable FV3JEDI_TEST_TIER=2. I'm not sure if re-running ecbuild is necessary, then you run make again.

The whole group of tier2 tests is pretty long to run but you shouldn't need to run the tier2 bump tests for hofx.

[HamidehGMAO]

Thanks for the instruction. I'll take a look.

[cmgas]

I see that the variable sss doesn't have a checksum associated with it, when the other ones have it. Does is matter much?

	double f10m(Time, zaxis_1, yaxis_1, xaxis_1) ;
		f10m:long_name = "f10m" ;
		f10m:units = "none" ;
		f10m:checksum = "C9624F96E8056332" ;

	double sss(Time, zaxis_1, yaxis_1, xaxis_1) ;
		sss:long_name = "sss" ;
		sss:units = "psu" ;

[HamidehGMAO]

Yeah! I don't think it does matter. Actually I don't know how the checksum value is generated for the other variables. I remember last time I added the sss to the bkg we decided to just ignore it.

[cmgas]

Thank you for re-adding these