Q&A – Hurricane Forecasting in the AI Age
I feature a listener question sent by one of these methods.
Email: brianmudd@iheartmedia.com
Social: @brianmuddradio
iHeartRadio: Use the Talkback feature – the microphone button on our station page in the iHeart app.
Today’s entry: Hey Brian – Love the show! I have a follow up question to your story about AI’s hurricane tracking success. You mentioned it was Google’s AI that was the most successful model last hurricane season. Are other AI companies working on hurricane models as well, and will the National Hurricane Center partner with these models/companies if they’re superior? After all, the point is the best information to keep people safe... Thanks!
Bottom Line: Today’s note references one of my Top 3 Takeaways from last Friday when I was talking about things that were better than ever. My second takeaway was that hurricane tracking has become more accurate than ever before because of AI, and specifically Google’s DeepMind AI, produced better results than the human-directed models. As mentioned... AI beat out the human programmed models at every interval – peaking with 10% more accuracy than any other model.
Among other things that will be in play as we approach this year’s hurricane season will be a still smaller cone than we’ve had before reflecting the increased confidence in hurricane tracking and forecasting in the AI age. But to your point there are so many different companies developing high level AI, is it the case that Google’s DeepMind model is the only one around? The answer is no.
On the back of DeepMind’s success, the National Hurricane Center is rolling out its own ‘suite’ of AI weather models with each having different applications. As outlined by the NHC – there will be three:
- AIGFS (Artificial Intelligence Global Forecast System): A weather forecast model that implements AI to deliver improved weather forecasts more quickly and efficiently than its traditional counterpart.
- AIGEFS (Artificial Intelligence Global Ensemble Forecast System): An AI-based ensemble system that provides a range of probable forecast outcomes to meteorologists and decision-makers. Early results show improved performance over the traditional GEFS, extending forecast skill by an additional 18 to 24 hours.
- HGEFS (Hybrid-GEFS): A pioneering, hybrid "grand ensemble" that combines the new AI-based AIGEFS (above) with NOAA’s flagship ensemble model, the Global Ensemble Forecast System. Initial testing shows that this model, a first-of-its kind approach for an operational weather center, consistently outperforms both the AI-only and physics-only ensemble systems.
The National Hurricane Center also provided these additional details about each of their new AI models and what the impact in detecting and projecting serve weather will be.
AIGFS — a new AI-based system that uses a variety of data sources to generate weather forecasts comparable to those produced by traditional weather prediction systems, such as GFS.
- Performance: shows improved forecast skill over the traditional GFS for many large-scale features. Notably, it demonstrates a significant reduction in tropical cyclone track errors at longer lead times.
- Efficiency: AIGFS’s most transformative feature. A single 16-day forecast uses only 0.3% of the computing resources of the operational GFS and finishes in approximately 40 minutes. This reduced latency means forecasters get critical data more quickly than they do from the traditional GFS.
AIGEFS — an AI-based 31-member ensemble, similar to the GEFS, that provides a range of possibilities for weather forecasters and decision-makers rather than a single forecast model solution.
- Performance: forecast skill is comparable to the operational GEFS.
- Efficiency: requires only 9% of the computing resources of the operational GEFS.
HGEFS — the most innovative application in the new suite. The HGEFS is a 62-member "grand ensemble" created by combining the 31 members of the physical GEFS with the 31 members of the AI-based AIGEFS.
- Performance: by combining two different modeling systems (one physics-based, one AI-based), the HGEFS creates a larger, more robust ensemble that more effectively represents forecast uncertainty. As a result, the HGEFS consistently outperforms both the GEFS and the AIGEFS across most major verification metrics.
- A NOAA first: to our knowledge, NOAA is the first organization in the world to implement such a hybrid physical-AI ensemble system.
2025 was the year that AI surpassed the human-directed forecasts – effectively providing proof of concept with Google’s model. This year is the year where AI will begin to be scaled not just by third party AI services like Google’s, but as a primary forecasting system by the NHC. In other words, we’ve reached the point where we’ll have hurricane forecasting to begin to be meaningfully reflected through the prism of AI.