ESM working group: Meeting notes 2024

Earth System Working Group
, ,
23

Date: 9/5/2024

Participants: 20

Chair: @eunpalim

Agenda:

1. Admin

@dkhutch:

  • Only met once so far.
  • Plan is for a similar structure to the last workshop.
  • Want to know what things went well, and any other ideas send to @dkhutch or working group chairs. Want to know how it went, and how it could be better.
  • ESM Working Group got a lot of value out of a working group meeting at last workshop. Have indicated would like another WG meeting at this workshop. Ocean and Land unlikely to have a WG meeting at the workshop. Potentially an opportunity to entrain some of the community who might otherwise attend other WG meetings.
  • Need a student representative for the organising committee. Please suggest someone else or yourself.

2. Shared experiments/ACCESS-NRI resources

  • ESM project lg87 has used 1.4MSU out of 1.9MSU allocation. We have asked, and received, two additional allocations of compute quota (total 1.1MSU extra).

Mostly used by @Dietmar_Dommenget and will use the remaining resources. There may be more resources available. @Dietmar_Dommenget could use another 1-2MSU if available.

@Dietmar_Dommenget will give a full report of the experiments in a presentation in the next meeting. Have done 3 experiments. Not yet analysed in depth, but seem to be working correctly.

3. Science talk

Spencer Wong: “Coupling a shallow water model to the UM atmosphere”

  • Couple a shallow water model (thermocline depth) to UM. Want UM interacting with thermocline anomalies the atmosphere model sees.
  • Aim is to couple a simple ocean model to a complex atmosphere, enabling idealised experiments
  • Shallow water model exists as a stand-alone implementation using a 1x1 degree between 51S and 51N. Using forced wind stress anomalies can reproduce ENSO behaviour.
  • Utilised an existing slab ocean implementation directly within UM, and replaced with shallow water model.
  • Many technical challenges: adding prognostic variables, grid differences, parallelisation issues
  • River routing model runs on a 1x1 degree grid, included prognostic variables, machinery for creating prognostic variables, serialising code and regridding routines. Copied and utilised this existing code.
  • Is a gather → calculate → scatter operation.
  • Performs much better than slab model
  • Peaks in Niño 3.4 autocorrelations are slimmer than in observations: El Niño/La Niña don’t persist as long as observed. More work required.
  • 2-3KSU for 50 year run. Throughput is 1-1.5 days per 50 years.

4. CMIP7 Update

@RachelLaw gave an update on plans for the ACCESS CMIP7 submission:

CMIP7-update-May2024.pdf (580.9 KB)

  • DECK is consistent with previous CMIP cycles with addition of historical. Note piControl and emissions-driven esm-piControl (interactive carbon cycle).
  • AR7 Fast Track is from a range of MIPs, but geared towards next IPCC report. Should be completed by end of 2026.
  • 6 scenarios. Recommendation to be emissions driven.
  • Thinking about feasible model development timeline. Still aiming for ESM3 for wider CMIP7. Hesitant about committing for fast-track.
  • ESM1.6 would derive cleanly from ESM1.5. Would be close to current configuration of CABLE but with some updates to land-use.
  • Thinking about a fast track submission with ESM1.6.
  • ESM1.6 1 degree ocean, CABLE3
  • ESM3 0.25 degree ocean, CABLE4
  • Don’t have resources for large changes to physics for ESM1.6. Do want to identify any outstanding issues or fixes.
  • The earlier the submission is in the more uptake from the community. Definitely want an ACCESS model in the fast-track. Don’t want to detract from ESM3.
  • Have not currently got the resources we would need for a more ambitious ESM3 on Fast Track timeline.
  • @JulieA Here’s a writeup from the recent CMIP meeting for those interested: https://wcrp-cmip.org/cmip-panel-meet-to-advance-cmip7-and-ar7-fast-track/
  • @Aidan Some bugs with Ocean BGC that need fixing. ESM3 should be done well, and not necessarily beholding to CMIP cycles. ESM3 might not run fast enough for Paleo work.
  • @RachelLaw scenario MIP is always higher priority and a lot of projection work flows from it. Community can run whatever MIP experiments were of interest.
  • @RachelLaw None of currently planned models go beyond N96 resolution because it is too expensive for CMIP7 but ESM3 would provide a code base for higher resolution work (e.g. for Weather of the 21st Century COE).

This is a wiki post , if you want to update any of the details in these meeting notes, or to add your own recollection of what was discussed then edit me rather than replying.

1 Like