Last week, Hurricane Isaias made landfall in the Florida Panhandle. Our thoughts are with everyone affected and with the recovery efforts still underway. Earlier and more accurate forecasting can help protect communities by giving them more time to prepare, and we are grateful that Google’s WeatherNext model was able to assist the @NWSNHC 's expert forecasters.

Based on initial evaluations, here's how our model supported the response to #Isaias:

- Tracking the Hurricane’s Path: WeatherNext was the most accurate model on the storm’s track. While traditional physics models predicted landfall closer to New Orleans, WeatherNext correctly predicted a path further east. This allowed NHC forecasters to highlight the potential for wind and storm surge impacts in the Florida Panhandle from their very first forecast.

- More Time to Prepare: Up to three days before landfall, our model consistently predicted its eventual location in the Florida Panhandle to within 60 km or less, supporting the NHC as they honed in on a track toward the Florida Panhandle and issued hurricane watches and warnings to give officials in Florida, Alabama, and Mississippi more time to prepare.

- Early Indications about Intensity: Early on, WeatherNext accurately predicted the intensification to a major hurricane. Alongside the regional hurricane models HWRF and HMON, WeatherNext’s intensity forecasts helped NHC forecasters to predict that the Isaias would approach the coast as a category 2 hurricane from the very first advisory.

Giving communities more time and more accurate warnings is critical for protecting lives and livelihoods. While model performance varies from storm to storm, Isaias reflects our broader findings. In our recent @Nature paper co-authored with NHC forecasters, evaluating 200+ global tropical cyclones, we found our model provides ~1 day of additional lead time for both track and intensity.

It’s been an honor for the @GoogleDeepMind and @GoogleResearch teams to work alongside NHC forecasters. By combining AI, physics models, satellite data, and the crucial expertise of meteorologists, we're excited to keep pushing the boundaries of forecasting to help protect communities.

It’s been an amazing few years for AI weather forecasting, and it was a pleasure to chat about it with @FryRsquared for the @GoogleDeepMind podcast.

We spoke about the real-world impact it’s having now, like our work with the National Hurricane Center, and forecasting of ell as the foundations: why weather is such a tough physics problem, the butterfly effect, and how AI models actually predict weather.

You can listen here:

Apple: podcasts.apple.com/gb/podcast/can…
Spotify: open.spotify.com/episode/2BvnGs…
YouTube: youtube.com/watch?v=O_EWbn…

I’m excited to share WeatherNext 3, our most powerful AI forecasting model yet.

WeatherNext 3 generates forecasts each hour and at high-resolution, giving timely and more localized predictions. It’s trained on new data sources, like live satellites, and achieves state-of-the-art skill, according to independent evaluations by Brightband.

Importantly, it will also be accessible to billions of people across Google products worldwide.

One of the features I’m most excited about is its ability to forecast precipitation, which was a longstanding challenge for global weather forecasting models. Precipitation forecasts can be up to 50% more accurate, with the greatest precipitation forecasting improvements in regions where forecasts have traditionally been less reliable.

There are also new variables that are particularly relevant for renewable energy forecasting, like turbine-height wind speed and high resolution cloud cover.

There are many other improvements across the board, and together this was the result of fantastic work by our teams at @GoogleDeepMind and @GoogleResearch .

Link: blog.google/innovation-and…

We're announcing a major update to Weather Lab, our interactive website for sharing Google’s AI weather models from @GoogleDeepMind and @GoogleResearch: deepmind.google.com/science/weathe…

Here's what's new:
🌎Global weather forecasts from WeatherNext 2 and MetNet
📅Historical forecasts up to 15 days ago
📍Point forecasts for a specific location
🌀View cyclone tracks alongside weather forecasts
💧Maps for temperature, precipitation, pressure and wind speed

We rely on accurate weather predictions for critical decisions - from supply chains to energy grids to crop planning. AI is transforming how we forecast weather.

Thrilled to share WeatherNext 2 - our latest work in groundbreaking forecasting technology!

Learn more: blog.google/technology/goo…

Now available on Earth Engine, BigQuery, and soon Vertex AI, WeatherNext 2 generates hundreds of weather scenarios 8x faster with hourly resolution, beating our previous model on 99.9% of targets.

Huge thanks to the @GoogleDeepMind and @GoogleResearch teams!

We’re using AI to improve cyclone prediction. 🌀

Introducing Weather Lab: a new interactive platform developed with @GoogleResearch, hosting our experimental AI weather model which can predict a cyclone’s track, intensity, size and structure.

Here’s how it works. 🧵

We're hiring a Research Engineer in AI for Sustainability @GoogleDeepMind (San Francisco / Mountain View).

Seeking strong engineers at the interface of machine learning, environmental sustainability, weather, dynamical systems, and/or remote sensing: boards.greenhouse.io/deepmind/jobs/…

🚨 WeatherBench Update

1. WeatherBench-X, our new evaluation code, is now on GitHub: github.com/google-researc…

2. New models (plus other small updates) on the WeatherBench website: sites.research.google/weatherbench/

1/n

We’re putting our most advanced AI for weather forecasting into more people’s hands. ☁️🌐

Scientists can now access @Google’s WeatherNext models as well as real-time and historical data to accelerate research, improve how we respond to disasters and much more. → goo.gle/3PW8LKS

Today we’re thrilled to announce that real-time and historical AI-based weather forecasts from @Google’s WeatherNext suite of models are now available on Earth Engine and BigQuery. Anyone can access and use these data for research, analysis and operational decision making, which we hope will help accelerate research and development in the weather and climate community.

You can check out the AI-powered forecasts here: deepmind.google/technologies/w…

GenCast is out in @Nature! It's the first high res ML ensemble weather forecast which outperformed the operational state of the art. And few more things have happened since the preprint was first released ⬇️🧵

Watch the moment that our President Professor Sir Jim McDonald revealed our 2024 #MacRobertAward winner @GoogleDeepMind at last night's #RAEngAwards Dinner. Find out more about their award-winning GraphCast innovation: raeng.org.uk/news/ai-weathe…

We’re thrilled to have won the prestigious #MacRobertAward for GraphCast - our AI model delivering 10-day weather forecasts with unprecedented accuracy in under one minute via deep learning and vast data sets.

Read more about the award here → raeng.org.uk/news/ai-weathe…

🏅 The winner of the 2024 #MacRobertAward is... @GoogleDeepMind

Huge congratulations to the team behind GraphCast, which uses machine learning and vast data sets to give highly accurate and timely weather predictions up to 10 days in advance: macrobertaward.raeng.org.uk #RAEngAwards

We're hiring a Research Scientist in AI for Sustainability @GoogleDeepMind (in Mountain View or Cambridge MA). Seeking experts in ML, environmental sustainability, weather, fluid dynamics, complex systems, combinatorial optimization, and/or remote sensing:
boards.greenhouse.io/deepmind/jobs/…

🌍 GenCast update! 🌍 New highlights include 0.25° resolution, predicting extreme weather, cyclones, and wind power production. Across all our evaluations, GenCast is better than the world’s top operational medium-range weather forecast

New paper: arxiv.org/abs/2312.15796…

🧵 1/8