Blanka Balogh

For me, all these questions are open. The challenges are exciting, but at the moment I feel lost. A lot to think about!

4. How to use PINNs, or should we use PINNs in climate modeling?

3. Related: How many models should we use to train AI models? One of the main strengths of the CMIP exercise is that it is a multimodel ensemble. If everyone uses the same dataset to train emulators (e.g. ERA5), will we be able to consider several emulators as a multimodel ensemble?

2. How to create the learning samples? Which parts should we emulate or improve with ML? Should we use observational data (if so, how?)? Since we don't have « observations » from warmer climates, we should also use model data (or at least physical constraints). But model data have biases.

This is a challenging issue that requires a lot of expertise in numerical modeling of the climate, which only a few people has worldwide. I’ve been using ARPEGE-climat since ~5 years now, but I think that this is not sufficient.
And things are changing fast, so it is difficult to make decisions.

1. How to adapt « legacy » Fortran codes to new hardwares? The DSL solution seems appealing (eg. Using GT4Py), but maybe using JAX in Python could be sufficient? Both options rely on packages that requires to be maintained (seems OK at the moment).

All this got me thinking about the use of ML/AI in climate science. In contrast to NWP, hybrid approaches still seem to be the best option. But there are tons of problems to solve, like:

Back here after a while. I had a wonderful beginning of the week in Zurich at the Exclaim! Symposium where I had a poster.
It really was amazing, the quality of the talks was GREAT and the people amazing! Many thanks and kudos to the organizers!

ArchesWeather: An efficient AI weather forecasting model at 1.5° resolution One of the guiding principles for designing AI-based weather forecasting systems is to embed physical constraints as inductive priors in the neural network architecture. A popular prior is locality, w...

🥇 ArchesWeather (Couairon et al., 2024).
Trained on ERA5 data using just 2 A100 GPUs for 2.5 day — an impressive achievement! This model, ArchesWeather, rivals other SoTA AI NWP models at 1.5° resolution, thanks to innovations in the attention layer.

arxiv.org/abs/2405.14527

🥈 Bano-Medina et al., Towards calibrated ensembles of neural weather model forecasts.
White the need to perturb model parameters can be debated, this paper tackles the challenge of sampling both model and input uncertainties in NN-based weather prediction.

essopenarchive.org/users/777909...

🥉 Hakim et al., Dynamical Tests of a DL Weather Prediction Model.
This short paper evaluates whether the dynamical behavior of PanguWeather aligns with expectations, by assessing the response of the model to small perturbations of the input.

journals.ametsoc.org/view/journal...

As 2024 comes to an end, here are my 3 favorite papers of the year on NN-based weather prediction. I’ve chosen ones that might not be on your radar but stand out for their originality or insights.

GraphDOP: Towards skilful data-driven medium-range weather forecasts learnt and initialised directly from observations We introduce GraphDOP, a new data-driven, end-to-end forecast system developed at the European Centre for Medium-Range Weather Forecasts (ECMWF) that is trained and initialised exclusively from Earth ...

With < 48 hours to go, ECMWF gets to claim this year's Christmas-time AI weather forecasting mic drop: arxiv.org/abs/2412.15687

I also read a lot and love sharing papers, websites, and GitHub repos I find interesting — something I hope to continue here.
Excited to connect with other AI and NWP/Climate enthousiasts!
3/3

Previously, I worked on efficient Fortran/Python coupling for a full GCM (ARP-GEM1) on heterogeneous HPC resources (using both CPU and GPU nodes at the same time). Now, I’m back to the AI side, focusing on sparse physics-informed neural networks for climate modeling.
2/3

Hi Bluesky! I realize I’ve never introduced myself. I’m a research scientist in the climate research group at Météo-France, where I focus on developing a hybrid global climate model that combines AI and physics based modeling.
1/3

Hi Ferran! You’ve just been added!

If you're training models with more than one loss term, I can again strongly recommend our ConFIG optimizer: tum-pbs.github.io/ConFIG/ , simply swap out Adam&Co. for ConFIG, and you can potentially see substantial reductions in your training loss 😁 We'd also be curious to hear how it works for you

Of course, you have been added!

Done!

ACE2: Accurately learning subseasonal to decadal atmospheric variability and forced responses Existing machine learning models of weather variability are not formulated to enable assessment of their response to varying external boundary conditions such as sea surface temperature and greenhouse...

The new ACE2 climate emulator from Oliver Watt-Meyer et al has very compelling results, with results that look comparable to NeuralGCM. Congrats to the AI2 team!
arxiv.org/abs/2411.112...

Haha I wish there was an option for that too. Thanks for sharing!

Here’s a starter pack for AI in Weather & Climate research! 🎉 I hope to see this grow over time. If I missed anyone, please let me know!

go.bsky.app/D6uzmRv

Back from ECMWF !

On recrute ! Nous recherchons un ingénieur de recherche en IA appliqué à la prévision numérique du temps, au CNRM.

CDD de 21 mois à partir du 01/04/2024, à Toulouse. Date limite de candidature : 05/01/2024.
emploi.cnrs.fr/Offres/CDD/U...

Hello Selorm, sorry, I don’t have any collaborators in Germany.

Hi, I am Blanka, and I am a research scientist using AI to make climate models more accurate. I enjoy discussing the use of AI in weather forecasting and climate, especially in atmospheric modeling.