user input exchange coefs

[bderembl]

Proposition for #77

[knutaros]

just for interest: is it possible to have different routines to calculate the exchange coefficients, but one central place where the fluxes are calculated, instead of having the flux calculations redundant for different methods?

…

On 12/8/25 14:36, Bruno Deremble wrote: Proposition for #77 <#77> ------------------------------------------------------------------------ You can view, comment on, or merge this pull request online at: #78 Commit Summary * 5efe82b <5efe82b> user input exchange coefs File Changes (5 files <https://github.com/gotm-model/code/pull/78/files>) * *M* src/airsea/CMakeLists.txt <https://github.com/gotm-model/code/pull/78/files#diff-918219028bd5899dd3a149d198073fa869fd09351c1bd1731014afe45317b289> (1) * *M* src/airsea/airsea.F90 <https://github.com/gotm-model/code/pull/78/files#diff-b92ae9d85fdaeee10bb0c95b3be743aadc900162ef5fb3136262102133bfe8a5> (12) * *M* src/airsea/airsea_fluxes.F90 <https://github.com/gotm-model/code/pull/78/files#diff-6bbbd7cf95318b6c0b9d22b88ffc65cbf0ab8d48b29bf8f327e4749c94d5bd3a> (2) * *M* src/airsea/airsea_variables.F90 <https://github.com/gotm-model/code/pull/78/files#diff-8bae0a34b3fbc3588b9dd7cf2f541b80f497749997fe08345f2c59621c05429e> (1) * *A* src/airsea/constant_cd.F90 <https://github.com/gotm-model/code/pull/78/files#diff-02b58e7339fa7c28c94bc4604411f58f2d3e536338adb9712d1c80cff2f0ac10> (136) Patch Links: * https://github.com/gotm-model/code/pull/78.patch * https://github.com/gotm-model/code/pull/78.diff — Reply to this email directly, view it on GitHub <#78>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AC2RV6STTZFTLU3XUKV4TCD4AV5FLAVCNFSM6AAAAACOMCHJZOVHI2DSMVQWIX3LMV43ASLTON2WKOZTG4YDMNBTG44DONI>. You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[bderembl]

It should be doable.

we will still have to recompute several quantities like (t_atm - t_ocn)

also there are several terms that modify the air-sea flux (not the drag): rain impact, webb correction.

we would need a 3 step process:

• compute Cds
• compute the fluxes
• correct the fluxes

[bolding]

Fairall does things slightly differently and will have to be modified as it does not have a 'clean' value for all 3 coefficients.

But we can still do it for Kondo and constant - will get rid of redundant code

[bolding]

maybe the cd's should be moved to airsea_variables - and we split out the actual flux calculations from Kondo - and this can then be re-used in a dedicated flux routine

still Fairall needs to be handled separately

[jornbr]

If major refactoring of the meteo code is desired, we might as well start from https://github.com/BoldingBruggeman/awex/ - which essentially is a refactored version of the GOTM meteo, adapted to also suit 3D models (pygetm). There the emphasis is on cd - but again fairall did not fit in easily, and is currently not enabled there.

[bolding]

What Jorn suggest makes a lot of sense :-)

I'll do that - over time ....

@bderembl - you can use your own implementation for now - right?

[bderembl]

great for me -- as long as there is a constant cd option in the new version ;)