What’s new at M²LInES: 🧊 FloeNet: ML emulator for sea-ice dynamics @willjgregory.bsky.social 🌐 Diffusion-based neural surrogate for stable climate simulations - Qi Liu, @joanbruna.bsky.social 🌎 Highlighted in the 2025 @schmidtsciences.bsky.social Impact Report mailchi.mp/4bca7e3e26ce...

🚨 We’re hiring at M²LInES! 2 postdoc positions on AI Earth System Models for seamless prediction, working with Laure Zanna, @joanbruna.bsky.social, Carlos Fernandez-Granda Apply: apply.interfolio.com/183889 apply.interfolio.com/183900 Please share! #AI #ClimateScience #MachineLearning

🌍 Announcing SamudrACE, our AI climate emulator built so scientists & planners can run “what-if” climate experiments quickly. Traditional models are slow and costly; SamudrACE makes high-quality simulations fast & more accessible. 🧵

Excited to share SamudrACE, the first 3D AI ocean–atm–sea-ice #climate emulator! 🚀 Simulates 800 years in 1 day on 1 GPU, ~100× faster than traditional models, straight from your laptop 👩‍💻 Collaboration with @ai2.bsky.social and GFDL, advancing #AIforScience with #DeepLearning. tinyurl.com/Samudrace

SamudrACE: A fast, accurate, efficient 3D coupled climate AI emulator

A fast digital twin of a state-of-the-art coupled climate model, simulating 800 years in 1 day with 1 GPU. SamudrACE combines two leading…

medium.com

✨Check out our new newsletter 👉 mailchi.mp/0608f769fe88... ✨ 🔹 SamudrACE – a new M²LInES–Ai2 ML coupled climate model 🔹 CAM5–CAM6 Supermodel – by William Chapman and colleagues 🔹 CERA – a new hashtag#machinelearning framework designed to improve climate-change generalization

September is here with a fresh M²LInES Newsletter! tinyurl.com/M2lines-sep25 🌀New neural net replacing Gent-McWilliams in ocean models 🔌From @leapstc.bsky.social, @yongquan-qu.bsky.social introduces Plug-and-Play DA 📰New IOP Journal - Machine Learning: Earth @pierregentine.bsky.social

Happy July! Our M²LInES Newsletter is out! mailchi.mp/8877fdc291c9... 🧊 @willjgregory.bsky.social et al. introduce a hybrid ML–climate model that improves sea ice forecasts + 🌊 Perezhogin et al. present a generalizable NN parametrization for mesoscale eddy. and more news!

We have a new @m2lines.bsky.social pre-print which shows how machine learning can be used for state-dependent bias correction to improve global coupled climate model sea ice forecasts - take a look! doi.org/10.48550/arX...

🌊 New in our June newsletter: 🧠 Verma et al. use deep learning to reduce surface ocean biases 🌐 Darman et al. boost ocean model generalization with transfer learning ➕ More updates & news! Read: mailchi.mp/105744018d4b... #OceanScience #ML #ClimateModeling

🚨 We’re hiring a postdoc at NYU Courant & @nyudatascience.bsky.social to build #AI foundation model for ocean & atmosphere 🌊☁️ as part of @m2lines.bsky.social! Work with Profs. Laure Zanna, @joanbruna.bsky.social & Carlos Fernandez-Granda. Apply: apply.interfolio.com/167915 #MLforScience #ClimateAI

🚀 The latest M²LInES newsletter is out! 🚀 🔹 The CAMulator is here: A machine-learned emulator of CAM6! 🧠 Advancing ensemble Kalman filters with deep learning ...and more! 📬 Check it out here: mailchi.mp/778fbbb17f80... #MachineLearning #ClimateScience #WeatherPrediction #MLforClimate

Novel technique to stabilize emulators, Sea-ice predictions, and the best future path for climate modeling: those are all in our M²LInES April newsletter and much more! ➡️ mailchi.mp/efd79c6eae08...

CDS researchers built Samudra, an AI ocean emulator that runs 150x faster than traditional models. Unlike past AI attempts, it stays stable for centuries. Developed at @m2lines.bsky.social by Laure Zanna, Carlos Fernandez-Granda, & others. nyudatascience.medium.com/samudra-the-...

Samudra: The AI Ocean Emulator That Runs 150 Times Faster Than Traditional Models

Samudra, an AI ocean emulator by CDS researchers, runs 150x faster than previous models — turning climate simulations from days into hours.

nyudatascience.medium.com

🚨 March Newsletter is out now! 🚨 mailchi.mp/14656511dff3... Check out our latest updates: 🔬 A new study by postdoc Alex Connolly on the power of Physics-Informed Machine Learning 🌊 How Machine Learning is used to improve sea surface height prediction and more!

Our February newsletter is out! Catch up on our latest news such as: our JupyterBook to learn machine learning for climate led by Dhruv Balwada, a recent study to address artifacts in ML parametrization, youtube channel by Carlos Fernandez-Granda, and more! mailchi.mp/382387c99943...

Postdoc Job Alert at Princeton University / NOAA GFDL puwebp.princeton.edu/AcadHire/app... If you're interested in sea ice modeling and machine learning, then look no further! This job is a great opportunity to be part of the exciting developments happening in the field of hybrid climate modeling

M²LInES is on Bluesky! What better way to start here than share all about our AI Global Ocean Emulator for Climate ? Samudra is stable, reliable, and 150x faster than a numerical model! 🌊 Read more about Samudra in our blog post👇 tinyurl.com/samudra-ai

Meet “Samudra”: M²LInES’ AI Global Ocean Emulator for Climate

🌊 M²LInES is excited to present Samudra (Sanskrit for Ocean) — a neural network global ocean emulator. Samudra is trained on data from a…

tinyurl.com