Use Bulk Formula or Net Flux Form

[yj7054]

Dear All,

I am confused about whether using the bulk flux or providing the net flux for the surface foring. I suppose this topic was discussed before but I didn't find it.

Do you have any criteria during selecting? Does it related with forcing frequency or ohter factors? What are the advantages or the disadvantages of the bulk flux (or the net flux)?

I suppose there is the negative feedback in the bulk formula which means the simulation doesn't deviate under the surface forcing, while the parameterization of the bulk formula may not robust compared with the net flux form?

Thank you.

[kate]

Think about say the Gulf Stream or the Kuroshio. If your fluxes were computed when it was wobbling south, colder water would be present and the fluxes would reflect that. If your Gulf Stream was instead wobbling north, the fluxes would therefore be off. Instead, using BULK_FLUXES, you get the fluxes based on your SST. Still, BULK_FLUXES use surface air temperature in a way that nudges your SST towards it.

[yj7054]

Think about say the Gulf Stream or the Kuroshio. If your fluxes were computed when it was wobbling south, colder water would be present and the fluxes would reflect that. If your Gulf Stream was instead wobbling north, the fluxes would therefore be off. Instead, using BULK_FLUXES, you get the fluxes based on your SST. Still, BULK_FLUXES use surface air temperature in a way that nudges your SST towards it.

Thank you!

You mean fluxes can reflect the state of the ocean, however the fluxes in BULK_FLUXES are decided by the air temperature.

Therefore, if we have robust air temperature data and we believe the SST should be close to the air temperature, we could turn on BULK_FLUXES to get better SST simulation. Otherwise, we would better use the fluxes, because it is what the ocean feels.

[kate]

I personally always use bulk_fluxes, in part because the sea ice model requires it.

If you don't use bulk_fluxes, it is advised to also use the dQdSST term to account for your SST differing from that used to compute the fluxes.

[yj7054]

Thank you very much!