Jason Stock delightfully surprised us with how much reasonable multiscale tropical rainfall variability can be generated with global autoregressive diffusion models trained only on satellite based rainfall products - a fun NVIDIA Earth-2 research internship result. Paper👇

This is our first step towards modernizing parameterization/data assimilation. We believe that differentiable programming can unlock new potentials in #AI4Science research. Looking forward to discussions at ICLR2024 workshop on AI4DifferentialEquations in Science!

Hear from the first customers using the Earth-2 cloud APIs, as they leverage #AI to accelerate global climate and weather simulations and visualization at a groundbreaking 2-kilometer scale. #GTC24

My colleagues at Pacific Northwest National Laboratory are seeking a postdoctoral scientist to help with the development of Machine Learning components of cloud physics schemes across scales from weather to climate.
Please see: careers.pnnl.gov/jobs/8751

A new published article, led by Oliver Watt-Meyer, investigates direct learning of heating/moistening rates from a coarsened 3 km (realistic topography!) storm-resolving simulation for use in a 200 km global climate model. 🌡🌧🌐
agupubs.onlinelibrary.wiley.com/doi/10.1029/20…

We live in such strange times. Apple, a company famous for its secrecy, published a paper with staggering amount of details on their multimodal foundation model. Those who are supposed to be open are now wayyy less than Apple.

MM1 is a treasure trove of analysis. They discuss

📢LEAP's Summer Programs are now open for applications!📢

💻Click for more details:
** 2024 REU (undergrads): bit.ly/3SPg1cs
** 2024 Momentum Fellowship (doctoral students): bit.ly/3UObHNb

#LEAPEducation #community #SummerIsComing
U.S. National Science Foundation Columbia Climate School Columbia Engineering

My team at Google is looking to hire a PhD student intern for research on AI-based coupled Earth system modeling. This would be a full-time ~3 month position in summer or fall 2024 working in-person in Cambridge, MA with Dmitrii Kochkov and janniyuval.

New from Earth2 research @NVIDIA: Intercomparing probabilistic skill across AI weather models can be hard. Noah Brenowitz has a crafty solution - lagged ensembles in Earth2MIP. Paper:  arxiv.org/abs/2401.15305. Code: github.com/NVIDIA/earth2m…. Or come to my AMS talk on Wed 1/31.

We're seeking a Postdoctoral Scholar with a passion for researching compound coastal flood risk assessment. Explore more details here:

recruit.ap.uci.edu/JPF08823

RT would be much appreciated.

It's been 50 years since the Global Atmospheric Research Programme Atlantic Tropical Experiment (GATE)!!

Want to learn more? Check out the GATE 50th Anniversary Celebration at the American Meteorological Society Annual Meeting: annual.ametsoc.org/index.cfm/2024…

#AMS2024

My warm & friendly academic home department UCI ESS UCI Physical Sciences UC Irvine is searching for a new atmospheric science assistant professor. We'll start reviewing applications in the coming weeks and it's not too late to apply. Please spread the word. recruit.ap.uci.edu/JPF08736

I’m at #AGU2023 near Poster 670 all morning. Swing by if you’d like to say hi or discuss ClimSim - our #NeurIPS2023 award winning benchmark dataset for hybrid physics- #ML climate simulation - fun big team project with LEAP-STC NVIDIA M²LInES Exascale Computing Project Allen Institute for AI & more.

Major congratulations to UCI ESS Dr Sungduk Yu + Prof Mike Pritchard +
colleagues for winning Best Paper Benchmark Award at the AI conference #NeurIPS for work at interface of AI and climate.

ClimSim: physics + ML for climate

LEAP-STC Exascale Computing Project

Meanwhile, NeurIPS Conference, ✨ClimSim won the Outstanding #Datasets & Benchmarks Track Award!✨

🎉 Congratulations to Sungduk Yu for taking the lead on this important paper! Hear him talk about #ClimSim on WEDNESDAY Dec. 13th @ 1:45p (CDT) + peep his poster from 3-5p (CDT). (1/4)

Thank you #NeurIPS for elevating our benchmark ClimSim with a Best Paper Award! A passion project spanning my roles at @UCI LEAP-STC, Exascale Computing Project & NVIDIA AI with Berkeley AI Research M²LInES & more. Sungduk Yu speaks on Wed am or find me at our poster #AGU2023 the next morning.

So great to see this emergence of a multi climate model benchmark dataset for training ML surrogates of the world’s vetted, government climate projections. Congratulations Julia Kaltenborn David Rolnick & co, excited to learn more!

I will be giving a talk on NeuralGCM (NG31A-06) at 9:25am on Wednesday morning at the session on 'Machine Learning Subgrid-Scale Parameterizations.'

Or come by at 3:30pm on Wednesday when I'll give the talk again at the Google booth.

Respect the strategic intent of this search UW Atmos Sci — great opportunity here for those on the faculty job market with expertise in AI and atmospheric science!

📢 Machine Learning faculty openings at UC Irvine 📢 This is a great year to apply since we seek to fill multiple positions in AI/ML. Join a great department in a fantastic location! recruit.ap.uci.edu/JPF08642