implement open water normalization for some ATM->OCN fluxes #44
Conversation
* add second conservative mapping type using aofrac normalization * normalized lwnet,sensible,latent and momentum fluxes from ATM by open water fraction in ATM when mapping to OCN
|
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 <#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
<#44 (comment)>, 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
|
|
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
The routehandle ( 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) |
|
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. |
|
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
<#44 (comment)>, 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
|
|
@SMoorthi-emc Can you please review/approve this PR so we'll be ready to merge once the ufs-weather RTs are done? Thanks. |
|
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. |
Description of changes
add second conservative mapping type using aofrac normalization. The RouteHandle for this new mapping type is a copy of the existing
mapconsfroutehandle for ATM->OCN. The sole purpose is to allow a second normalization type using the open water fraction from the ATM. See related issue single normalization type is not enforced #40 which enforces a check against using the same routehandle for more than one normalization type.normalize lwnet,sensible,latent and momentum fluxes from ATM by ATM's open water fraction when mapping to OCN
Specific notes
Fixes Issue #29