Diagnose the cause for degradation in results with sea ice insertion

[sanAkel]

Issue is described in the ⬇️ figure.

Needs further investigation:

• How is it that without any insertion (left panel), areas with almost 0.0 (white) OMF now shows up as (colored) in the middle panel? Please pick a point/region, zoom into it and plot time-series.
• While the ice edges seem to improve, a time series of the hemisphere averaged OMF will help further diagnostics.
• Is there a plot of SST and SSS? And their differences from the SST (v2.5)? Similarly SSS(v2.5)?
• With a time-series plot of the mean and standard deviation of the increments in the: a. Arctic, b. Antarctic.

[zgarraffo]

The left panel didn't have 30 days but only 1 day SIC with no insertion. CICE6 is very close to the initial condition. Will check.

[sanAkel]

The left panel didn't have 30 days but only 1 day SIC with no insertion. CICE6 is very close to the initial condition. Will check.

Can you please update:

• The figure with 30 days (or as many as done)?
• Also plot other figures.

Thanks!

[zgarraffo]

I don't have figures for 30 days for the free ice run. I did 1-day for experiment 14.3, will put it in. In the presentation the first slides should be changed from expt 14.2 (no ice assimilation) to 14.3 (with ice assimilation). Please confirm if I can do that.

…

On Mon, Oct 13, 2025 at 8:09 AM Santha Akella ***@***.***> wrote: *sanAkel* left a comment (NOAA-EMC/RTOFS_GLO#85) <#85 (comment)> The left panel didn't have 30 days but only 1 day SIC with no insertion. CICE6 is very close to the initial condition. Will check. Can you please update: - The figure with 30 days (or as many as done)? - Also plot other figures. Thanks! — Reply to this email directly, view it on GitHub <#85 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AZFCD47QONAV65CYPBDLRUL3XOI5ZAVCNFSM6AAAAACI76UHEWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTGOJXGI2TMOBZHA> . You are receiving this because you were assigned.Message ID: ***@***.***>

[zgarraffo]

I don't have a 30 day figure for expt 14.2 (I use the ncoda verification and no ice concentration was created). Expt_14.3 is the ice verification that I have.

[zgarraffo]

Confirmed that the 1st day ice coverage average is identical for expt_14.2 (initial RTOFS cice4 converted to cice6) and for expt_14.3, which will have ice assimilation. This first day is before assimilation. So omit expt_14.2 1-day ice coverage. I want to move expt_14.3 to the top part of the presentation.

[sanAkel]

Confirmed that the 1st day ice coverage average is identical for expt_14.2 (initial RTOFS cice4 converted to cice6) and for expt_14.3, which will have ice assimilation. This first day is before assimilation. So omit expt_14.2 1-day ice coverage. I want to move expt_14.3 to the top part of the presentation.

Please do add a new slide with the figure(s) you propose.

[zgarraffo]

I will move expt_14.3 to the top of the slides.

[zgarraffo]

Santha, please do the polar plots that you wanted.
Monthly mean netcdf files of SST and SSS analysis increments, in ursa
MOM6 expt_14.3:
/scratch3/NCEPDEV/marine/Zulema.Garraffo/FV3_RT/expt_14.3/ncoda
seatmp_surf_202504_mean_analinc_14.3.nc and
salint_surf_202504_mean_analinc_14.3.nc
RTOFS:
/scratch4/NCEPDEV/marine/Zulema.Garraffo/EMC.rtofs.v2.5.a/ncoda
seatmp_surf_202504_mean_analinc_EMCv2.5.a.nc and
salint_surf_202504_mean_analinc_EMCv2.5.a.nc

[sanAkel]

Global plots for SST (top) and SSS (bottom) from the ⬆️ mentioned path/files

@zgarraffo would it be possible to make a time-series of the mean and standard deviation (aka RMSE) that is shown in the titles of the above plots?

[zgarraffo]

time series plots of the mean and standard deviation?

Yes. (1)

Or time series of a horizontal mean variable day by day?

No.

It will have to wait some.

Please provide an ETA for above (1).

Were you going to do polar plots of the monthly mean SST and SSS?

Yes. ✅

Note: The path(s) you gave ⬆️ have netcdf output files for 1-day only. If you point me to the corresponding monthlies, it can be done. Until then, plotting for:

/scratch3/NCEPDEV/marine/Zulema.Garraffo/FV3_RT/expt_14.3/ncoda
seatmp_surf_202504_mean_analinc_14.3.nc and
salint_surf_202504_mean_analinc_14.3.nc
RTOFS:
/scratch4/NCEPDEV/marine/Zulema.Garraffo/EMC.rtofs.v2.5.a/ncoda
seatmp_surf_202504_mean_analinc_EMCv2.5.a.nc and
salint_surf_202504_mean_analinc_EMCv2.5.a.nc

[zgarraffo]

Files like /scratch3/NCEPDEV/marine/Zulema.Garraffo/FV3_RT/expt_14.3/ncoda
seatmp_surf_202504_mean_analinc_14.3.nc
are the monthly mean, 20250401 through 20250430. No timeseries can be plotted for a time mean ( I didn't put a correct time in the netcdf file) You want timeseries of errors over the globe or for different regions? I can easily produce a time series of global mean SST errors. I have to dig into the ncoda analysis programs to produce the same for SSS. I can start with global mean SST time series. I can check if I can produce timeseries of SST errors for regions.

[sanAkel]

@zgarraffo Please

I can start with global mean SST time series.

🙏

What we (hope) to see:

• Mean (m) and standard deviation (std) of the increments in any variable over time asymptote.
• A system, if well tuned, m and std are ⬇️, certainly not ⬆️-- ⬅️ Good to check that!

m and std in titles of these figures show that we have work to do!