Adding new output variable of USWRFItoa

[XiaSun-Atmos]

Hi,
I am trying to add a new output variable of USWRFItoa, instantaneous top of atmos upward shortwave flux. It is defined in GFS_diagnostics.F90. This variable is added to diag_table:
"gfs_phys", "USWRFItoa", "uswrf_toa", "fv3_history2d", "all", .false., "none", 2
However, the output values of uswrf_toa from fv3 runs are in magnitudes of 1e6 W/m2, which should be in magnitudes of 1e3 W/m2. The spatial distribution does look very similar to uswrf_avetoa, which is one hour avg of uswrf_toa.
What should I do to make uswrf_toa on the right magnitudes?
Thanks,

[yangfanglin]

Does the surface DSWRFI look fine ?

There might be an issue. All the TOA instantaneous SW and LW fluxes are passed from fluxr in ccpp/driver/GFS_diagnostics.F90; However, fluxr in ccpp/physics/physics/GFS_rrtmg_post.F90 are accumulated variables. They are not instantaneous fluxes.

Surface instantaneous fluxes look to be fine.

Bring the attention to
@dustinswales @grantfirl @Qingfu-Liu @junwang-noaa

[XiaSun-Atmos]

Thanks @yangfanglin . All the other instantaneous radiation fluxes that use their designated variable names, such as dswsfci(:) and ulwsfci(:) (not fluxr) look good.
cc: @joeolson42

[XiaSun-Atmos]

Hi, I am following up with this discussion.
Is there a plan or timeline for fixing the code for the TOA instantaneous SW and LW fluxes?
Thanks.

[yangfanglin]

@XiaSun-Atmos , We at EMC do not have a plan to work on this yet. It would be great if you or anyone else in the community can fix this bug.

[RatkoVasic-NOAA]

@yangfanglin Is this issue solved?

[XiaSun-Atmos]

I don't think so, unless some other developers have fixed it.