COSIMA TWG Meeting Minutes 2026

COSIMA TWG
, ,
12

6 May 2026

Andy, Helen, Anton, Ezhil , Chris, Siobhan, Manodeep, Dougie, Kieran, Andrew

Liquid runoff spreading (Anton)

Anton suggest fixed (in time) remapping weights which spread selected high volumne rivers only, and map all the rest of the runoff directly to coastal cells

We probably want latent heat extracted for basal heat - does that play in?
This probably needs mixing at depth for basal melt

  • should we be doing this ?
  • we probably think this is a fair bit of work ?

Spreading increases with grid resolution - the higher resolution the more the spreading is needed.
River fluxes are quite seasonal - we actually only need the spreading for a few months of the year
Spreading exponentially doesn’t represent rivers that well - in reality the freshwater plume follows the coast (depending on the hemisphere ?)

Siobhan - Why is the basal scheme important ? Is this SOFIA protocal ?
Andy - this would add another file dependency that means yet another file to generate at new configs
Dougie - we probably don’t want another scheme with a fixed jump in pattern as the month’s change

Issue: Runoff redistribution · Issue #396 · ACCESS-NRI/access-om3-configs · GitHub

MOM Dev Meeting (Dougie)

NSAA/NOAA/NCAR/ACCESS all have their own CMake build systems. These orgs will meet seperately to see if a shared CMake build is possible

  • likely hard to coordinate this into a single effort

Possibly next PR will change answers for KPP configs (due to a change in other packages)

OM3 Optimisations (Andy)

MOM6 - NON-blocking updates (Manodeep)

  • Why does non blocking updates break restart reproducability (i.e. over different run lengths)

  • We are also surprised about why does this change answers ? it implies a bug in the code where it useing some memory before it has been updated by an MPI call

8km global (Chris)