Flux corrected projections with ESM1.5

Earth System
1

Hi All,

I am in the process of setting up a flux corrected version of ESM1.5. I am following the experimental set up from this paper (https://essopenarchive.org/doi/pdf/10.22541/essoar.15005957/v1?download=true&redirectToLatest=false), as this will be part of a larger multi-model assessment.

The flux correction is done in a couple of stages, the first one being to run the model while restoring SST near globally to ERA5 SSTs for the period 1950-1979 (this is almost complete, thanks to discussions and assistance from Ariaan, Georgy, Spencer). After this, the next phase is to apply a surface wind stress correction that is calculated as the wind stress difference between the SST restored simulation and those from ERA5 (Then there is a final stage of applying a surface temperature correction, but we don’t need those details here).

The reason for this email is I am wondering if anyone has experience applying a correction to modelled wind stresses before they are applied to the ocean? Where/how is the best place to introduce this in the code?

Regards, Shayne

2

Hi Shayne,
I’ve never done it myself, but I have been recently messing around with the part of the code that does the flux corrections. Specifically, in the namelist ocean_sbc_nml, there is a flag called do_flux_correction, located here:

And if you want to do a wind stress correction, you need to give it the input files with naming convention as per this part of the code:

If the input directory sees that file called tau_x_correction.nc and the do_flux_correction flag is set to .true., then it will read it in the file and apply that correction field. The same applies for

tau_y_correction.nc
temp_sfc_correction.nc
salt_sfc_correction.nc

which are also found in that same module.

NB: you don’t have to worry about making a folder called INPUT. That is taken care of by payu. You just have to specify the files in your config.yaml and then make sure that in ocean/input.nml you have switched on the flux correction flag.

3

I am almost certain @MATTENGLAND’s group did some flux correction experiments back in the day. I can’t recall details but maybe @PSpence or @rmholmes can remember who did that work?