CM3 Dev-Eval Working group meeting minutes
Meeting link
Date: Tuesday 20 January 2026 at 10:30 AM AEDT
Participants: @cbull, @wghuneke, @Dietmar_Dommenget, @kieranricardo, @MartinDix, @Harun_Rashid, @paocorrales, @jemmajeffree, @ctychung, @aekiss, @ariaan, @sofarrell, @ezhilsabareesh8, @paulleopardi
Chair: @kieranricardo
Minutes: @spencerwong
Agenda:
- Reminder about program committee nominations for ESM/CMIP7 focussed workshop in September.
- shout-out props for Figures from: @jemmajeffree @ctychung @wghuneke
- CM3 technical meeting updates (Dec 17th, …)
- offer support (breakout room?) to people wanting to get involved / could set aside 10 minutes at end of meeting to answer questions.
Program committee nominations for CMIP7/ESM focussed workshop in September
- ACCESS-NRI will be hosting a CMIP7/ESM focused workshop in early September in Melbourne. We are currently putting together a program committee to help plan the workshop. We are looking for people who would bring valuable expertise, perspectives, and commitment to shaping the workshop’s program. We also want to build a diverse committee that includes representation across career stages and institutions.
- If you are interested in being part of the program committee, please contact @spencerwong or the co-chairs for the ESM and Land working groups by January 22, 2026.
Updates from CM3 technical meetings
Minutes from previous technical meetings are available here.
- @kieranricardo has been working on updating component versions for CM3’s ocean to bring them into line with OM3. The UM is also being updated to version 13.8 to allow for the full GC5 central configuration to be used.
- Tests with the UM13.8 and full GC5 central configuration showed sudden cooling, and took cloud forcing results further away from observations. Results will be compared with AMIP simulations from the UKMO.
- A new run will be done once the updates are completed, and this will also include improvements to the wind stress remapping which will likely have large influences on the model’s ENSO.
- There are still issues with excessive ice volume in the Northern Hemisphere. A similar problem exists in OM3 due to non-advective points in the grid, however the distribution of excessive ice in CM3 is quite different, indicating a seperate problem.
GitHub issue links:
-
@jemmajeffree ENSO and IOD power spectra & seasonal cycle re-visited
The Niño3.4 autocorrelation suggests too strong a 4 year cycle, while the ENSO amplitude is also much too high. -
@ctychung time series of CM3 (blue), 0.25 CM2 (green) and obs (orange)
Timeseries of the Niño3.4 anomaly show that a very strong El Niño and La Niña both occur in the first 10 years of the CM3 run. Plots omitting the first 10 years show that the Niño3.4 standard deviation remains too high even without these events. -
@wghuneke OM3 vs CM3 Bottom age
OM3 shows improvements in DSW overflow compared to OM2. While CM3 also has some improvements in the DSW overflow compared to CM2, however it also has significantly excessive open ocean convection. This could be related to model spinup, and we’ll need to check how it looks later in a longer run.
Community figures
Comparing CM3’s ENSO statistics with CM2 and the CMIP6 ensemble. Two statistics that stand out are the thermocline depth and the thermocline depth standard deviation, both which are much higher than observations
Please paste here and include a caption!