SST/SICs with OSTIA in RNS producing erroneous sea ice concentrations

Atmosphere Regional Nesting Suite
1

Hi Team and in particular @cbengel,

I’ve done my first run with the OSTIA sst/sic turned on for my Antarctic domain, but unfortunately I seem to have run into some issues. Attached are three images:

Screenshot 2026-01-19 at 13.54.09
Screenshot 2026-01-19 at 13.54.09675×717 80.7 KB

Screenshot 2026-01-19 at 13.54.01
Screenshot 2026-01-19 at 13.54.01699×679 68 KB

Screenshot 2026-01-19 at 13.53.54
Screenshot 2026-01-19 at 13.53.54717×714 72.5 KB

the ancillary produced by OSTIA suite (top) and two (nested) domains from my run where you can see that there is a large problem with the sea ice concentrations :frowning: This error seems to occur when the initial conditions are generated as the/scratch/jk72/slf563/cylc-run/u-di850/share/cycle/20240117T0000Z/Regn1/resn_2/RAL3P2_glomap/ics/ostia_seaice.anc files show the same thing with just a constant near 1. Any ideas where to look to fix this?

Thanks!

2

Hi @sonyafiddes!

Thanks for opening this topic – super cool to see you working on this for Antarctica! We’ll find the best person to provide some support on this and get back to you here in the next day or two!

Cheers,

Lawrence

3

Thanks - I had a thought that clearly something has gone wrong with the re-mapping - could this be caused by my use of a rotated polar grid?

4

Hi Team, just giving this a bump here - it would be great to see if we can find a solution!

5

@sonyafiddes sorry for the wait. Nobody knows the RNS like Chermelle, but she is on leave unfortunately. So it is taking us a lot longer to get an idea of what might be happening, that we could share with you. Unfortunately, you may have to wait a bit more (especially with the upcoming NCI maintenance), but we haven’t forgotten your issue.

6

Hi @sonyafiddes, let’s make a time and discuss this offline. I wonder if there may be some problem with units?

Would it be possible to catch up after AMOS?? Or do you have time-pressing commitments?

7

Hi Chemelle,
Yes - after AMOS would be great :slight_smile: Thanks!

8

Hi Chermelle,

As discussed - I’m going to try and hack a solution for this, but it woul dbe good to get to the bottom of it.

I’m pretty free on Tuesday next week - can we chat then?

(PS. I thought I had found a nother problem, but I think I was wrong - so have modified this post!)

9

Absolutely. Let’s chat. I will arrange a time with you offline.

10

Hi @sonyafiddes and @cbengel - just wanted to check on the status of this! Was this issue resolved offline? If so, if you think it has a solution that might be helpful to the community, it would be great to summarize in a post here. :blush: If you would rather not share the solution for any reason, that’s no problem, and I can just close this topic as is.

If this conversation is ongoing and you’d like to leave this topic open, please reply in the next 2 weeks (by May 4). If I don’t see a reply by then, I’ll go ahead and close this topic. You are always encouraged to open a new topic if you have any further questions at a later time. Thanks! :blush:

11

Hi Paige,

We did not resolve this problem. I have a work around now that is not very neat (I take the global OSTIA SSTs and regrid them myself, using mule). I think the problem is within the OSTIA package itself which we think is not something ACCESS-NRI can really help with, probably need to enlist help of the MetOffice… which I didn’t really want to wait for…

So - I guess I would go ahead and close this, and if I ever want to solve it properly, then I can open a new ticket!

Cheers,

Sonya

12

Thanks for your quick response @sonyafiddes and for summarizing what you have learned about this issue! I’ll go ahead and mark your most recent post as the “solution” so that anyone who comes across this thread will see your workaround, and then I’ll close the topic.

Please do open another topic if you end up coming back to this issue at some point in the future. :blush: