GOCART AMIP Develop only runs on certain dates: SU volcanoes

[mathomp4]

Updated issue

The Ops folks managed to copy over all the files, so that is a good thing. Ops emissions runs (where explosive is /dev/null) works.

However, there is now another issue. AMIP runs are currently limited to dates where explosive files exist.

For example, I tried to run AMIP on 2000-04-14 but the model crashes:

forrtl: severe (41): insufficient virtual memory
Image              PC                Routine            Line        Source
GEOSgcm.x          0000000003964A2B  Unknown               Unknown  Unknown
GEOSgcm.x          000000000393BC33  Unknown               Unknown  Unknown
GEOSgcm.x          0000000000E64924  gocart2g_process_       10471  GOCART2G_Process.F90
GEOSgcm.x          000000000092282F  su2g_gridcompmod_         864  SU2G_GridCompMod.F90
GEOSgcm.x          00000000009165FF  su2g_gridcompmod_         698  SU2G_GridCompMod.F90

My guess is that it is looking for a file for that day but none exists as there are only 19 files for the year 2000:

> ls -1 *2000*
 so2_explosive_volcanic_emissions_CARN_v202401.20000128.rc
 so2_explosive_volcanic_emissions_CARN_v202401.20000129.rc
 so2_explosive_volcanic_emissions_CARN_v202401.20000130.rc
 so2_explosive_volcanic_emissions_CARN_v202401.20000131.rc
 so2_explosive_volcanic_emissions_CARN_v202401.20000201.rc
 so2_explosive_volcanic_emissions_CARN_v202401.20000203.rc
 so2_explosive_volcanic_emissions_CARN_v202401.20000205.rc
 so2_explosive_volcanic_emissions_CARN_v202401.20000206.rc
 so2_explosive_volcanic_emissions_CARN_v202401.20000207.rc
 so2_explosive_volcanic_emissions_CARN_v202401.20000208.rc
 so2_explosive_volcanic_emissions_CARN_v202401.20000210.rc
 so2_explosive_volcanic_emissions_CARN_v202401.20000211.rc
 so2_explosive_volcanic_emissions_CARN_v202401.20000226.rc
 so2_explosive_volcanic_emissions_CARN_v202401.20000320.rc
 so2_explosive_volcanic_emissions_CARN_v202401.20000818.rc
 so2_explosive_volcanic_emissions_CARN_v202401.20000828.rc
 so2_explosive_volcanic_emissions_CARN_v202401.20000928.rc
 so2_explosive_volcanic_emissions_CARN_v202401.20001215.rc
 so2_explosive_volcanic_emissions_CARN_v202401.20001219.rc

If I edit the SU2G instance file to just hardcode in, say, the 20000818 file no matter the date, things run (I'm guessing a continuously exploding volcano might make a weird climate, but it runs).

I believe SU needs some sort of protection if an explosive file does not exist for a given date.

---

Old issue

The below was fixed by copying over files into main ExtData location.

Pinging @amdasilva and @pcolarco

Currently in develop we have for Ops emissions:

# Volcanic pointwise sources
volcano_srcfilen_explosive: /dev/null
volcano_srcfilen_degassing: ExtData/chemistry/CARN/v202401/so2_volcanic_emissions_CARN_v202401.degassing_only.rc

and for AMIP:

# Volcanic pointwise sources
volcano_srcfilen_explosive: ExtData/chemistry/CARN/v202401/explosive/so2_explosive_volcanic_emissions_CARN_v202401.%y4%m2%d2.rc
volcano_srcfilen_degassing: ExtData/chemistry/CARN/v202401/so2_volcanic_emissions_CARN_v202401.degassing_only.rc

There are two issues here.

1. The degassing path is incorrect and needs a sfc/ directory after the v202401: ExtData/chemistry/CARN/v202401/sfc/so2_volcanic_emissions_CARN_v202401.degassing_only.rc
2. The explosive files do not exist in the default ExtData area. I see them in $SHARE/dasilva/fvInput/ExtData/chemistry/CARN/v202401/explosive but they are not in $SHARE/gmao_ops/fvInput/ExtData/chemistry/CARN/v202401/

[mathomp4]

Oh wait. Maybe we don't need to fix this but need to organize files correctly. Let me reach out to the ops folks.

[pcolarco]

Matt, let me ask at our group meeting today. From: Matt Thompson ***@***.***> Date: Tuesday, October 8, 2024 at 10:02 AM To: GEOS-ESM/GOCART ***@***.***> Cc: Colarco, Peter R. (GSFC-6140) ***@***.***>, Mention ***@***.***> Subject: [EXTERNAL] [BULK] Re: [GEOS-ESM/GOCART] GOCART Develop currently does not run: SU volcanoes (Issue #293) CAUTION: This email originated from outside of NASA. Please take care when clicking links or opening attachments. Use the "Report Message" button to report suspicious messages to the NASA SOC. Oh wait. Maybe we don't need to fix this but need to organize files correctly. Let me reach out to the ops folks. — Reply to this email directly, view it on GitHub<#293 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ANW73YH44GDCUU6C5JVHRF3Z2PQW3AVCNFSM6AAAAABPSJEOM2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDGOJZHE2DKMJYHE>. You are receiving this because you were mentioned.Message ID: ***@***.***>

[mathomp4]

@pcolarco et al

An update. So, the Ops groups copied over the files and the good news is Ops emissions can run without changes. The problem now is AMIP doesn't run on some days. I'm updating the main comment above about this.

[jrjoshi1]

FYI @pcolarco and others: my test for GOCART-develop ran without crashing for the tested one day, after adding a control to check the volcanic-file's existence. BUT the code did not read any of the volcanic files (regardless their existence) perhaps because the 'fname' was not defined (e.g., here

GOCART/ESMF/GOCART2G_GridComp/SU2G_GridComp/SU2G_GridCompMod.F90

Line 851 in ffa44c7

call StrTemplate(fname, self%volcano_srcfilen_degassing, xid='unknown', &

). Will be testing by defining fname ...

[jrjoshi1]

update: defining 'fname' in the code worked, the model now read in the degassing files (the test was with AMIP emissions for 2000-04-14)

[mathomp4]

update: defining 'fname' in the code worked, the model now read in the degassing files (the test was with AMIP emissions for 2000-04-14)

@jrjoshi1 Once you have a PR, let me know and I'll test it as well. Thanks!

[jrjoshi1]

@mathomp4 I have a PR now: #294. Thanks.