search

geoschem / geos-chem

GEOS-Chem "Science Codebase" repository. Contains GEOS-Chem science routines, run directory generation scripts, and interface code. This repository is used as a submodule within the GCClassic and GCHP wrappers, as well as in other modeling contexts (external ESMs).
http://geos-chem.org
Other
169 stars 165 forks source link

[BUG/ISSUE] Is MULTISEAICE supposed to be 1984-2017? #1158

Closed LiamBindle closed 2 years ago

LiamBindle commented 2 years ago

What institution are you from?

WUSTL

Description of the problem

I noticed that MULTISEAICE in HEMCO_Config.rc has a period of 1984-2017 https://github.com/geoschem/geos-chem/blob/9ab6931db5b556eafd0e002e9fd4fdbcbd5fbb44/run/GCHP/HEMCO_Config.rc.templates/HEMCO_Config.rc.fullchem#L2890 but http://geoschemdata.wustl.edu/ExtData/HEMCO/MULTI_ICE/v2021-07/ has data for 1980-2020. Should the period in HEMCO_Config.rc actually be 1980-2020? Any idea @msulprizio @yantosca or @lizziel?

IIUC, GC-Classic is using the 2017 file for 2018 onwards.

(fyi @laestrada)

LiamBindle commented 2 years ago

Same goes for CEDS_NO_SHP in CEDS/v2021-06. The data period in HEMCO_Config is 1750-2010 but the data in v2021-06 is 1750-2019. https://github.com/geoschem/geos-chem/blob/9ab6931db5b556eafd0e002e9fd4fdbcbd5fbb44/run/GCHP/HEMCO_Config.rc.templates/HEMCO_Config.rc.fullchem#L2415

CEDS/v2021-06 on data portal: http://geoschemdata.wustl.edu/ExtData/HEMCO/CEDS/v2021-06/

Shoudl the line in HEMCO_Config be 1750-2019?

msulprizio commented 2 years ago

That's definitely a typo for CEDS_NO_SHIP. It should be updated to 2019 for the latest version of CEDS. We can fix that in 13.4.0. I believe this will change benchmark output because HEMCO is likely using the wrong year (2017) for the benchmark simulations when it should be using 2019.

msulprizio commented 2 years ago

@lizziel It looks like you added the MULTI_ICE files, likely for the emissions from blowing snow update? I assume the years should be 1980-2020 as @LiamBindle points out. Can you confirm?

lizziel commented 2 years ago

Yes, that date range was an error. The HEMCO_Config.rc submission for the update in https://github.com/geoschem/geos-chem/pull/356 contained 1984-2017, and then the developer said she was submitting data for 1984-2018. I never checked the actual data submitted to change that string. Yes that will change the benchmark.

msulprizio commented 2 years ago

Thanks for confirming. I'll go ahead and fix both typos before we proceed to benchmarking. Thanks again for catching this @LiamBindle!

LiamBindle commented 2 years ago

Awesome, thanks @msulprizio and @lizziel!

yantosca commented 2 years ago

We can close this issue since PR #1159 has been merged into the 13.4.0 development stream. At this time 13.4.0 is pending benchmark approval.