So excited to share our Google DeepMind team's new Nature paper on GenCast, an ML-based probabilistic weather forecasting model: www.nature.com/articles/s41... It represents a substantial step forward in how we predict weather and assess the risk of extreme events. 🌪️🧵 - ThreadSky

tom-andersson.bsky.social • 89 days ago

So excited to share our Google DeepMind team's new Nature paper on GenCast, an ML-based probabilistic weather forecasting model: https://www.nature.com/articles/s41586-024-08252-9

It represents a substantial step forward in how we predict weather and assess the risk of extreme events. 🌪️🧵

Comments

tom-andersson.bsky.social•89 days ago

GenCast uses diffusion to generate multiple 15-day forecast trajectories for the atmosphere. It assigns more accurate probabilities to possible weather scenarios than the SoTA physics-based ensemble system from ECMWF, across a 2019 evaluation period.

tom-andersson.bsky.social•89 days ago

It’s vital that we ensure these new ML weather systems are safe and reliable. One thing I'm proud of is our range of evaluation experiments: per-grid-cell skill & calibration, spatial structure, renewable energy, extreme cold/heat/wind, and the paths of tropical cyclones (i.e. hurricanes).

tom-andersson.bsky.social•89 days ago

For example, we created a dataset of simulated wind power data at wind farm sites across the globe, and found that GenCast outperforms ENS by 10–20% up to 4 days ahead. This is promising, because better weather forecasts can reduce renewable energy uncertainty and accelerate decarbonisation.

tom-andersson.bsky.social•89 days ago

We also extracted cyclone tracks from GenCast and ENS and compared them with ~100 cyclones observed in 2019. GenCast's ensemble mean cyclone track has a 12-hour position error advantage over ENS out to 4 days, and more actionable track probability fields out to 7 days.

tom-andersson.bsky.social•89 days ago

Cyclone max wind speeds are still underestimated, but this performance on tracks is really promising.

One recent devastating cyclone was Hurricane Milton, which caused >$85 billion in damages. GenCast predicted ~70% probability of landfall in Florida 8.5 days before it struck.

tom-andersson.bsky.social•89 days ago

A GenCast ensemble member takes 8 minutes on a TPU chip, versus hours on a supercomputer for physics-based models. This opens up the possibility of large ensembles (eg 1000s of members), which could better estimate risks of extreme events. We don't yet know how much this will help.

kylecranmer.bsky.social•89 days ago

Congratulations!

Comments

Posting Rules

Reply