How should monthly SOLARIS-HEPPA solar forcing be extended beyond 2299?

[mdeushimri]

We would appreciate clarification on how to extend the monthly SOLARIS-HEPPA solar forcing beyond 31 December 2299 for the (esm-)scen7-x-ext experiment.

The current guidance specifies that the forcing should be repeated from 24 August 2038, with 1 January 2300 corresponding to 24 August 2038. However, our atmospheric chemistry model uses a proleptic Gregorian calendar and accepts only the officially provided monthly SSI data, which are used to calculate photolysis rates.

Because the repetition begins in the middle of a month, the specified daily mapping cannot be reproduced exactly using the provided calendar-month means alone.

Could you please clarify the formally recommended and CMIP7-compliant procedure for models that use only the provided monthly data?

We would prefer not to generate new monthly data from the daily dataset, as the averaging and calendar-mapping procedure could introduce arbitrary methodological choices. Our preference is therefore to use the officially a provided monthly dataset without additional temporal aggregation.

In this case, should January 2300 use the August 2038 monthly value, followed sequentially by September 2038 for February 2300, or is another procedure recommended?

[znichollscr]

@berndfunke @st-bender could you help here please? Thanks!

[st-bender]

Hi,

We had a little internal discussion about it, the better way would be to use Sep 2038 instead of Aug 2038 for Jan 2300 to concatenate the monthly data, see figure. However, this can still result in differences of about 0.5 W/m^2 in the TSI when compared to monthly averaging the extended daily data.

Please note that I am not aware of a future calendar change and proleptic_gregorian will probably be used for a while and there should be no confusion in that regard. Thus producing the monthly-averaged data from the daily series is straightforward, setting the time coordinate to the 15 of each month. However, to avoid any confusion, we have produced an extended data set for those who are interested:

https://cloud.iaa.es/index.php/s/Qej32YyNkrCkBMG

@znicholls Feel free to upload this data set along side the standard scenarios. I tried to keep the data and descriptions consistent and updated for 230001-256104, please let me know if I missed something that should be updated to match your consistency criteria.

[victoriarspada]

Hello @st-bender I notice now that this discussion is a duplicate of the discussion I started here #491 . Has/will the extended file you attached here be uploaded to ESGF? I cannot seem to access the link you attached here, and it would be great to access the file from a more permanent location. Thanks!

[znichollscr]

Sorry this slipped off my radar. I'll get this onto ESGF asap. In the meantime, here's a link which should work as a temporary workaround if you're in a rush https://drive.google.com/file/d/1c1tz32xU45PnRqiuJHrDWnKfauO7yoMU/view?usp=drive_link

[znichollscr]

The files is now published on ESGF, but there's an issue with ORNL so you can't find it/get it yet. It should be up soon. In the meantime, use the link above

[mdeushimri]

Thank you very much for your detailed response and for taking the time to prepare the extended dataset. For our model, we plan to use the extended monthly dataset that you have prepared based on the daily data. We greatly appreciate your support.

[mdeushimri]

We checked the extended dataset and noticed that the variable wlenbinsize(wlen) appears to be missing. If possible, we would appreciate it if you could add this variable to the extended dataset as well.

[st-bender]

We checked the extended dataset and noticed that the variable wlenbinsize(wlen) appears to be missing. If possible, we would appreciate it if you could add this variable to the extended dataset as well.

Thanks for the note, that is indeed a mistake, we will update the file in due time.
Meanwhile, you can use/copy the missing variables from the usual monthly data set as they are independent of time and did not change.

Best.

[st-bender]

Hello,

I have updated the file and it should now be complete (same link).
Let me know if there is something else missing.

Best.

[mdeushimri]

Thank you for providing the corrected data. I have reviewed it and did not find any issues.
I appreciate your prompt response and support.

Best,