mediterranean sea salinity

[anton-seaice]

Hi @har917 - could you share the plot of eastern vs western sea surface salinity in the Mediterranean please?

Did the change which reduced Nile river outflow go into cableUM7/esm1.6 part way through the spin-up runs? If so which year of the run ?

Thanks

(ps. This is related to interest in seeing some plots of the amount of through flow between the eastern and western Mediterranean in esm1.6. The eastern part is getting very salty and its not clear if it's related to runoff or ocean mixing.
)

[MartinDix]

ACCESS-Community-Hub/access-esm1.6-dev-experiments#4 (comment)

[anton-seaice]

So the change went in at the start of the "April" spin-up ?

e.g.

/scratch/p66/pjb581/access-esm/archive/pi_concentrations_test3-restart-840-35bbe5a7 is "before"

and

/scratch/p66/jxs599/access-esm/archive/AprilSpinUp-AprilSpinUp-56aae814 is "after" 

?

[har917]

@anton-seaice good @MartinDix remembered which issue/repo we uploaded this work into ....

Our understanding is that the reduction in run off originated due to a bug-fix that went into the UM executable when we combined to the new executables in April (specifically the UM uses the namelist value for cmt-opt consistently through out the code). This is a UM change (and not CABLE/land). The relevant section(s) of the Hive forum post are last two currently there (so years 0-235 and years 236-current which are outputting into /scratch/p66/jxs599/access-esm/archive/AprilSpinUp-AprilSpinUp-56aae814)

@JhanSrbinovsky @pearseb @blimlim could you check whether the detail in the forum post is correct - it indicates that the exe is pr31-3 which (as it's the same as earlier sections) is likely incorrect. We possibly need to note any dependencies on payu changes in that forum post as well

[anton-seaice]

Thanks Ian - FYI @ezhilsabareesh8

[JhanSrbinovsky]

Updated details in forum post mentioned above.

@anton That bug fix to Convective Mass Transport is in UM7 repo @pr43

The change went in just before Xmas. Pearse's run just missed out on this change. @dougiesquire and @blimlim verified that Pearse's executable dates back toDec 5th. We totally missed it in testing until the ESM had enough years done to validate looking at SSS.

[anton-seaice]

Thanks @JhanSrbinovsky

There is a question about these lines in the forum post:

Years: 236 - present (=338 @april 30th)
Exe versions: access-esm1p6/pr31-3

Where I think the exe should be access-esm1p6/dev_2025.04.000 ?

[JhanSrbinovsky]

I think I made that change @anton-seaice - maybe I didnt save the edit, but yes it should be. the PR31-3 is actually the same code but pre-release

[JhanSrbinovsky]

done now