Stephen Yeager (Nov 02 2020 at 22:48): Stephen Yeager (Nov 02 2020 at 22:48):All, Here is a link to the SMYLE output spreadsheet:
https://docs.google.com/spreadsheets/d/1BQ7f_OrtZog3rnlXWhzIFODaRZkjpBAnFWe1rWpo9Ds/edit?usp=sharing
Please suggest edits and/or discuss below.
Fred Castruccio (Nov 04 2020 at 00:28):Daily winds at 10 meters? This will allow to perform dynamical regional downscaling. Since no other dataset (as far as I know) includes all the variables needed to do a clean downscaling, it might be a real plus in my view.
Stephen Yeager (Dec 15 2020 at 22:25):@Keith Lindsay @Matt Long We're planning to save POP budget terms for the following tracers: DIC, TEMP, SALT.
The full tracer budget requires these 3D fields: HDIFB_, HDIFE_, HDIFN_, KPP_SRC_, UE_, VN_, WT_, + any SGS advection and/or source terms. Could we economize on output by saving _TEND and computing mixing (HDIF + KPP) as a residual? This would allow us to eliminate in net 9 3D fields.
Keith Lindsay (Dec 22 2020 at 03:12):Steve, I apologize for not getting back to you earlier, I've been swamped.
I'm not clear on which terms you are proposing dropping.
Note that the model has a variety of processed based tendency terms available.
I think that these were not used in the CESM2 LENS because Keith Rodgers wanted fluxes, not tendencies.
TEND_TEMP RF_TEND_TEMP ISOP_ADV_TEND_TEMP Redi_TEND_TEMP SUBM_ADV_TEND_TEMP HDIF_EXPLICIT_3D_TEMP ADV_3D_TEMP
Stephen Yeager (Dec 22 2020 at 17:42):Keith, Column L of the SMYLE output spreadsheet lists a bunch of 3D fields that adds significantly to the data footprint. I'm still looking for about ~10GB/year to save by eliminating some of these. I'm proposing dropping the following (for VAR={DIC,TEMP,SALT}): DIA_IMPVF_VAR, HDIFB_VAR, HDIFE_VAR, HDIFN_VAR, KPP_SRC_VAR. I'd add the following: TEND_VAR. This would still allow for budget analysis, but not a detailed breakdown of mixing/diffusion terms.
Last updated: May 16 2025 at 17:14 UTC