search

NOAA-EMC / CMEPS

NUOPC Community Mediator for Earth Prediction Systems
https://escomp.github.io/CMEPS/
0 stars 20 forks source link

implement open water normalization for some ATM->OCN fluxes #44

Closed DeniseWorthen closed 3 years ago

DeniseWorthen commented 3 years ago

Description of changes

Specific notes

Fixes Issue #29

SMoorthi-emc commented 3 years ago

Denise, I don't think I really understand what is going on in CMEPS. I used to know a little bit in NEMS/Mediator. I am not even sure what this "normalization" means and what is projected from what to what. Thanks Moorthi

On Wed, May 5, 2021 at 3:36 PM Denise Worthen @.***> wrote:

@DeniseWorthen https://github.com/DeniseWorthen requested your review on: #44 https://github.com/NOAA-EMC/CMEPS/pull/44 implement open water normalization for some ATM->OCN fluxes .

— You are receiving this because your review was requested. Reply to this email directly, view it on GitHub https://github.com/NOAA-EMC/CMEPS/pull/44#event-4690989541, or unsubscribe https://github.com/notifications/unsubscribe-auth/ALLVRYWMJ72SDQB27CEVZ5LTMGM3BANCNFSM432DAEXA .

-- Dr. Shrinivas Moorthi Research Meteorologist Modeling and Data Assimilation Branch Environmental Modeling Center / National Centers for Environmental Prediction 5830 University Research Court - (W/NP23), College Park MD 20740 USA Tel: (301)683-3718

e-mail: @.*** Phone: (301) 683-3718 Fax: (301) 683-3718

DeniseWorthen commented 3 years ago

Hi Moorthi,

The purpose here is to normalize the fields from the ATM which are calculated only over the (time-varying) open water fraction, analogously to how the fluxes from ICE->ATM are normalized by the time-varying ice-fraction.

There are 5 fluxes that I understand are exported from ATM->OCN which are "open-water only" fluxes: u,v momentum, lwnet, sensible and latent. In esmFldsExchange_nems_mod.F90, the normalization by open water area is implemented when mapping these ATM fields to the OCN, for example the lwnet:

call addmap(fldListFr(compatm)%flds, 'Faxa_lwnet', compocn, mapconsf_aofrac, 'aofrac', 'unset')

The routehandle (mapconsf_aofrac) is "map conservative fraction using aofrac as the normalization" (aofrac is the open-water fraction from the ATM). The name of the normalization field to use is aofrac.

In the NEMS mediator, this sort of operation was implemented similarly, for example in the prep_atm when the ice fraction was mapped to the ATM, the field from the ice was multiplied by ice fraction and mapped to the ATM, and then the mapped field was multiplied by the reciprocal of the mapped ice fraction. In CMEPS, this happens in the med_map_field_normalized (LN1102)

DeniseWorthen commented 3 years ago

Actually, a clarification: Even when the open-water fraction does not vary in time (for example, non-ice covered areas), the fluxes mapped to the OCN will still be normalized by the fraction of open water on the ATM grid (the mapped ocean mask). So, this will impact fluxes along coastal areas outside of ice-covered regions.

SMoorthi-emc commented 3 years ago

Hi Denise, Thanks for the explanation. However, I still have some confusion. Suppose we ignore ice fraction for now. i.e. water fraction is fixed in time. With the fractional grid, near the coast, FV3 water fraction is going to be less than one. However, on the ocean grid, the water fraction is always 1. I am concerned that if the fluxes from ATM are weighted by water fraction, the coastal ocean points may get reduced fluxes? Suppose the coastal point is full water in the ocean model, but half land and half water in ATM. Then what would be the appropriate flux (e.g. radiative flux) to the full ocean grid? Moorthi

On Thu, May 6, 2021 at 8:28 AM Denise Worthen @.***> wrote:

Actually, a clarification: Even when the open-water fraction does not vary in time (for example, non-ice covered areas), the fluxes mapped to the OCN will still be normalized by the fraction of open water on the ATM grid (the mapped ocean mask). So, this will impact fluxes along coastal areas outside of ice-covered regions.

— You are receiving this because your review was requested. Reply to this email directly, view it on GitHub https://github.com/NOAA-EMC/CMEPS/pull/44#issuecomment-833482797, or unsubscribe https://github.com/notifications/unsubscribe-auth/ALLVRYRQFYOJG64XCAMASJ3TMKDO7ANCNFSM432DAEXA .

-- Dr. Shrinivas Moorthi Research Meteorologist Modeling and Data Assimilation Branch Environmental Modeling Center / National Centers for Environmental Prediction 5830 University Research Court - (W/NP23), College Park MD 20740 USA Tel: (301)683-3718

e-mail: @.*** Phone: (301) 683-3718 Fax: (301) 683-3718

DeniseWorthen commented 3 years ago

@SMoorthi-emc Can you please review/approve this PR so we'll be ready to merge once the ufs-weather RTs are done? Thanks.

DeniseWorthen commented 3 years ago

Adding both @MinsukJi-NOAA or @binli2337 for review. Both Jun and Moorthi have approved, but since Moorthi doesn't have write permissions I need one of the two of you to review/approve. Thanks.