Jesse Ferrell retweetledi
Flash Flood Warning including Pearl City HI, Kailua HI and Waipahu HI until 3:30 PM HST
English
Jesse Ferrell
49.8K posts
Jesse Ferrell
@WeatherMatrix
Jesse Ferrell, AccuWeather Meteorologist & Digital producer. Opinions my own. WeatherMatrix, Since 1996: "The Weather is Always Interesting."
State College, PA Katılım Mayıs 2008
1.5K Takip Edilen10K Takipçiler
Jesse Ferrell retweetledi
How often do you get tornadoes without hail or straight-line wind reports? Turns out, quite frequently. "All hazards" only occur near tornadoes about 45% of the time.
"Tornado only" risk is most common in the Southeast and Midwest, and "Tor+hail" most common in the High Plains.
English
Jesse Ferrell retweetledi
Jesse Ferrell retweetledi
I have finally completed my capstone study at the University of Oklahoma. Lots of really interesting findings and analysis on my poster!
English
After a self-driving Waymo robotaxi drove into floodwaters and was swept away into a creek last month, the vehicles have been recalled for a software upgrade.
accuweather.com/en/weather-new…
English
Jesse Ferrell retweetledi
On this day in 1997: an F1 tornado with estimated winds of 100-110 mph went through downtown Miami and into Biscayne Bay before lifting and touching down again in Miami Beach
Video via: WSVN
English
Jesse Ferrell retweetledi
Well, we can say that the season is officially open!
May 11th 2026.
Istria, Croatia
#supercell #hail #storm #severeweather
English
Jesse Ferrell retweetledi
🌡️ Up to 45 °C in northern India
Temperatures in northern India are rising to as high as 45 °C, and several-day heat waves, locally severe, are expected this week across parts of northwest, western, and central India.
India Meteorological Department (IMD) has issued a warning for severe heat wave conditions over parts of western Rajasthan. ⚠️
View current and forecast temperatures on Windy.com, including station observations from the past 10 days.
English
Jesse Ferrell retweetledi
What an absolutely stunning sunrise this morning from Cherry Grove Beach. The colors are stunning and mesmerizing! Simply beautiful Rick Farrell.
English
Jesse Ferrell retweetledi
Jesse Ferrell retweetledi
Check out the impressive shelf cloud with the storm over Myrtle Beach right now.
English
Jesse Ferrell retweetledi
[5/11/26 3:00 PM]
Severe weather threat ENDED for eastern NC, with some stronger storms now along our coastal waters. Take a look at this picture of a waterspout earlier between Harkers Island and Marshallberg taken from Tom Roberts!
#ncwx #waterspout
English
Interesting if true
Yohan@yohaniddawela
Physics-based weather models still beat AI when it matters most. Not on average. On the most extreme days. This is the opposite of what we've been hearing... A new paper in Science Advances ran every major AI weather model: GraphCast, Pangu-Weather, Fuxi, against ECMWF's HRES across 162,751 record-breaking heat events, 32,991 cold records, and 53,345 wind records in 2020. On average conditions, the AI models win. GraphCast, Fuxi, and the rest outperform HRES on standard temperature and wind benchmarks across most lead times. This matches what every prior benchmark study has shown. AI weather forecasting is genuinely impressive. Then the researchers asked a different question. What happens when the event is unprecedented? Not extreme. Not the 95th percentile. Actually beyond anything in the training data. HRES won every single category. Heat records. Cold records. Wind records. Nearly every lead time. The performance gap was largest at short lead times, where AI models should have the most information and the least uncertainty. The bias pattern is pretty massive. The AI models systematically underestimated how extreme the events were. The bigger the record exceedance, the larger the underprediction. The researchers describe it as an implicit 'soft cap': the models behave as if they can't forecast values much beyond the most extreme thing in their training data. The bias grows almost linearly with how far the event exceeded the record. HRES showed no such pattern. This isn't a fluke. The same result held in 2018 and 2020, which had opposite ENSO conditions. It held across the tropics, subtropics, mid-latitudes, and high latitudes. It held for all three variables. It held when the researchers ran an alternative evaluation specifically designed to avoid the forecaster's dilemma. The mechanism is pretty straightforward. AI weather models are trained on ERA5 reanalysis data from 1979 to 2017. They learn to interpolate between historical weather patterns. When a new initial condition arrives, they find the nearest analogues in training and produce something in between. Record-breaking events, by definition, have no close analogues. The model has never seen anything quite like this, so it regresses toward the most extreme things it has. Physics-based models like HRES don't work this way. They solve partial differential equations describing atmospheric dynamics. They don't need a historical analogue for a 48°C heatwave in Siberia. The physics doesn't care whether it's happened before. The authors are careful about what this means. AI models remain faster, cheaper, and competitive on average conditions. Probabilistic AI forecasting is developing rapidly. Data augmentation with simulated extreme events and hybrid physics-AI architectures are plausible paths forward. This isn't a verdict on AI weather forecasting broadly. But the policy implication is quite important. The events where AI models fail hardest are exactly the events where accurate forecasting matters most. Record-shattering heat. Unprecedented wind storms. The scenarios that overwhelm emergency response, strain infrastructure, and kill people because no one expected them to be that bad. The authors wrote it plainly: it remains vital to fund and run physics-based NWP and AI weather models in parallel. I find it an unusually direct recommendation in a methods paper. Climate change means record-breaking events are becoming more frequent, not less. The training distribution is shifting. AI models trained on 1979 to 2017 data will see more and more out-of-distribution events as the climate diverges from that baseline. The extrapolation problem the researchers identified isn't going away. It's getting harder. The models that can't forecast records are being asked to forecast a world that's setting them constantly. Link to full paper: science.org/doi/10.1126/sc…
English
Jesse Ferrell retweetledi
Here's a look at the last week's power outages over the US. Highlights included weather-driven outages from a wet snowstorm in Colorado, tornadoes in Mississippi, and thunderstorms in Texas.
GIF
English
Then I realized I've taken a LOT of Asperitus photos, so I made a gallery. #slide-1889779" target="_blank" rel="nofollow noopener">accuweather.com/en/gallery/asp…
English
Did an article on Asperitus clouds yesterday. These were sighted in Springfield, Missouri on April 28. accuweather.com/en/weather-new…
English
7 tornadoes confirmed in southern Mississippi area from May 6th evening. Two are EF3s. accuweather.com/en/severe-weat…
English
Jesse Ferrell retweetledi
Jesse Ferrell retweetledi
Neural networks might speak English, but they think in shapes.
Understanding their rich *neural geometry* is key to understanding how they work – and to debugging and controlling them with precision.
Starting today, we’re releasing a series of posts on this research agenda. 🧵
English