Need to dial-in instead? Enter the meeting ID: 894 5980 026[date-range from=2023-05-25T14:00:00 to=2023-05-25T15:00:00 timezone=âAustralia/Sydneyâ]8 via +61 3 7018 2005 or +61 2 8015 6011
Reminder
Our first Forecasting and Prediction working group meeting will be this Thursday at 2-3pm (Melb., Hobart, Sydney time)
Please email me an introduction slide if youâd like to use one! Also okay if you just want to introduce yourself and what youâd like to get out of this working group.
Disclaimer: This meeting is hosted by the Bureau of Meteorology. You must join with your full name and follow instruction given by the host. Disclosure or unauthorised use of content from this meeting may be a serious criminal offence.
We are running a 1-day workshop following the main ACCESS-NRI workshop.
TOPIC:Machine learning for modelling and prediction WHEN: Thurs 7 September; 9:30 AM - 4:00 PM WHERE: Canberra or online
The day will provide a brief introduction to ML, explore recent progress in the application of ML in modelling and prediction, and include a discussion session on key issues and future opportunities. This session is open to all, and no prior experience with, or knowledge of ML is required.
Register now (before 20 August). More details are provided in the registration link.
Membership of the Forecasting and Prediction WG is not required for attendance. Feel free to circulate the announcement.
We are looking forward to the FPWG Workshop next week Thursday (online and in Canberra) on Machine learning for modelling and prediction.
If you cannot attend the workshop, you are very welcome to include your thoughts on our Discussion Jamboard. We will be using the Jamboard to share and discuss ideas.
Our main topics of discussion are:
ML research interests and collaboration opportunities
How can ACCESS-NRI support (data, software, infrastructure, resources/info/guides)?
How can we better coordinate across the community?
Increasing ML research transparency, increasing collaboration and open source code
⊠but there is also a page for any other questions/issues.
The link to the Jamboard is live and we encourage anyone to add their ideas/thoughts at anytime â prior or during the workshop. Apart from facilitating our discussion session, we hope that it will be a useful resource for capturing information and will contribute to informing the outcomes of the workshop.
Disclaimer: This meeting is hosted by the Bureau of Meteorology. You must join with your full name and follow instruction given by the host. Disclosure or unauthorised use of content from this meeting may be a serious criminal offence.
Disclaimer: This meeting is hosted by the Bureau of Meteorology. You must join with your full name and follow instruction given by the host. Disclosure or unauthorised use of content from this meeting may be a serious criminal offence.
Aidan
(Aidan Heerdegen, ACCESS-NRI Release Team Lead)
17
Reminder: Shared ESM/F&P Working Group Meeting
This is a reminder the next ESM/F&P Shared Working Group meeting is on tomorrow, Thursday 29th of February at 1:00pm (2024-02-29T02:00:00Z).
This is a shared meeting between the ESM Working Group and the Forecasting and Prediction Working Group. The meeting will be split into two sections, the first covering the shared material, and the second dealing with ESM specific items, which all are welcome to stay for, but we also understand if F&P folks prefer to leave at that point.
We would very much welcome your comments/additions to the document. Please use the google-docs version for adding your suggestions: How ACCESS NRI could support ML/AI
The contents of the document are useful for the NRI in considering future investment options.
Joint Meeting: Atmosphere + Forecasting and Prediction WGs 08.05.2024
Weâve organised another Forecasting & Prediction WG meeting jointly with the Atmosphere WG. This meeting will focus on some shared modelling interests with the single column model configuration, and lidar and radar simulators. Please come and share your ideas for collaborative projects!
** For a calendar invite, simply click on the relevant calendar link below **
Single Column Model (SCM): presentation by Martin Dix (15 + 5mins)
The UM includes a SCM configuration. This includes the full UM atmosphere and land surface physics, but replaces the dynamics by advective tendencies derived from observations or from a full model. Itâs a useful tool for testing physics parameterisations because it runs quickly and gives detailed per timestep diagnostics.
Martin will present some tests comparing JULES and CABLE land surface schemes driven with data from ERA5 and from a global UM run.
Earth Model Column Collaboratory (EMC2): Calum Knight will give an introduction to EMC2, an open-source ground-based lidar and radar instrument simulator and subcolumn generator for large-scale models (10 + 5 mins)
Discussion: Ideas for future collaborative projects and joint meetings for Atmosphere and Forecasting & Prediction WGs (15 mins)
