Meeting Minutes 2023: Atmosphere Working Group

Atmosphere Working Group
, ,
12

Atmosphere WG Meeting 22.02.2024 minutes
22 Feb 2024
15 participants on Zoom
From: UNSW, BoM, Monash, ACCESS-NRI, CSIRO, UTas, UKMO

Matt Woodhouse:

  • Welcome & Acknowledgment

Chermelle Engel – Tutorial on STASH

  • We are talking about the ACCESS model, which is a complex model with a lot of details
  • Writing output needs to be carefully considered with respect to time stepping and variables. This is a very complex task and also depends on particular needs (eg. forecasts, reanalysis, climate scale)
  • In ACCESS this done via STASH (cudos to those who built it!) which is very flexible but modification needs to be done carefully
  • Pre-defined STASH packs provide default settings that would meet the majority of users (hopefully).
  • In the RNS there are five default STASH packs, plus the option to set your own either in the GUI or by creating your own STASH packs.
  • Creating your own STASH packs gives you flexibility to turn on and off your outputs as you need it (eg. for a period of particular interest).
  • Example of Canberra hailstorm where Chermelle creates new STASH profiles and requests (watch the recording).
  • Chermelle recommends turning on really high-res stuff only for the specific period you are interested in

James Warner, UKMO, visiting the BoM: ‘K-Scale: Scale interactions within the Tropical Easterly Jet’

  • Comparison of two configurations over the tropics RAL3.9 and GAL9 in the UM at 10km resolution
  • Representation of regional scale convection may impact at the global scale; an up-scale impact
  • Tropical Easterly Jet has important implications for the Indian and West African Monsoon
  • GAL9 produces are very large jet, larger than ERA5, RAL3 produces a closer jet to ERA5
  • When you have consistent physics with the driving model and the regional model you get similar results to the parent model, mixing physics (between driving model and regional model) can introduce differences
  • In GAL9: Weaker convection, poleward branch of Hadley Cell is weaker, altering the jet
  • Improving convection could help global models resolve issues with Indian Monsoon
  • James is here for another week if you have more questions

Heidi Nettelbeck: ECR and student engagement:

  • Can we encourage students and ECRs to attend these meetings
  • Let Heidi know if you have ideas in how to get people more engaged
  • Matt suggests the words ‘working group’ could put people off

Community updates

Belinda roux (BoM)

  • BARPA-R are available on the NCI, BARRA R2/RE2 is also available
  • ACCESS-AE is under development now, as is the data assimilation for ACCESS-A
  • BARPA-C and BARRA-C are also under development
  • Joint BoM-RND/UM workshop on convective scale modelling 9-13th September

Matt Woodhouse (CSIRO)

  • We have a year to do model configuration, tuning and evaluation before deck experiments must start.
  • Next FY model testing and tuning (long simulations & spin up), documentation
  • 25/26 FY deck experiments must begin
  • Time is VERY short
  • Right now no one is funded to do any CMIP7 dev… Rachel Law is trying to coordinate this, including large support from ACCESS-NRI and NCI. A lot of science needs to come from BoM and Universities.
  • Ideally, CM3 and ESM3 will have new physical climate model (UM13x), new ocean model (MOM6), new sea ice model (CICE6) and new land surface (CABLE4)
  • Unfortunately, for CMIP6 no tuning was done for ACCESS atmosphere… this was not ideal so hopefully we can do better this time around
  • Get involved: ACCESS Evaluation Hackathon is happening in a few weeks – this is a good way to get involved. Follow the CMIP7 announce topic CMIP7 Announce
  • See recording for info-graphics
  • The fast-track proposal is currently open for comment for CMIP7, if you would like to contribute there.
  • Fresh-eyes in CMIP is a network of ECRs to work with CMIP.
  • See recording for info-graphics / links

Steve Sherwood (UNSW)

  • W21C is in process of advertising a tranche of new postdocs, ads should be appearing soon.

Heidi Nettelbeck: Storage and Compute

  • We have 875kSUs and 95TB of storage to use this quarter. Post on the forum if you would like to utilise some of this. Start a New Topic in the Atmosphere Experiment sub-category and it will automatically populate the topic with a template.

Next meetings:

  • Should we have a fixed time?
  • Next one will probably be late March – date TBD
  • Interest in doing a joint meeting with the forecast & prediction WG on the single column model, likely in April
  • Agenda is open to all to contribute!
2 Likes