Introducing WeatherNext 3, our most advanced and accurate global weather AI model

From Google

 
Our flagship AI weather forecasting model now includes real-time satellite data, hourly refreshes, higher resolution, precise precipitation forecasting, and clean energy variables.
It’s now integrated across Search, Gemini, Maps, Google Maps Platform, and Cloud.

Every day, the weather influences billions of decisions.
Some are as simple as grabbing an umbrella before heading out the door, but others are far more consequential.
Wind, rain, and extreme weather events, like heatwaves and droughts, have cascading impacts across agriculture, global supply chains, clean energy production, and national economies.

In recent years, AI has revolutionized weather forecasting, using historical records to make faster and more accurate predictions than traditional methods.
Yet predicting highly local and rapidly changing weather has remained a challenge.
Previous models often lacked sufficient spatial resolution, and struggled to incorporate real-time weather data from sources like satellites.

Today, Google DeepMind and Google Research are introducing WeatherNext 3, the most advanced and accurate global weather model to date, according to independent live evaluations by Brightband.
Our model learns directly from real-time observations, enabling it to provide timely and more localized predictions for the weather events that impact people the most.
By using raw satellite data to produce a forecast every hour in high resolution, our model makes reliable forecasts accessible across Google products worldwide.

Predicting the weather is one of the oldest, most complex challenges we face.

Traditional numerical weather prediction models are powerful, but they can be slow to run and costly at a global scale.

WeatherNext 3 takes a different approach.

Built on a fundamentally different architecture, WeatherNext 3 learns directly from real-world observations, including live satellite feeds and ground-level weather station data.
What makes WeatherNext 3 different?

Hourly refresh: While traditional models typically refresh every six hours, WN3 produces a fresh forecast every single hour.

Hyper-Local resolution: Native 5km resolution for temperature and humidity, enabling better detail for coastal regions, mountains, and urban areas.

Actionable data: Includes variables like 100m wind speeds for wind energy management and detailed cloud/radiation metrics for solar energy planning.

Global reach: Delivering high-resolution insights everywhere.

You can now use WeatherNext 3 through Google Search, Gemini, Google Maps and more.

Rapid weather prediction at unprecedented resolution

A forecast's utility often comes down to detail and how finely it resolves both time and space.
WeatherNext 3 generates hourly forecasts at multiple spatial resolutions, maintaining physical consistency from broad global wind patterns all the way down to local topography.

With WeatherNext 3, we can visualize key surface variables — like temperature and moisture — at a 5-kilometer resolution, other surface variables at 10 kilometers, and atmospheric variables, like wind speed, at 25 kilometers.
Overall, this provides a global weather picture roughly five times sharper than our previous model, WeatherNext 2, which produced forecasts on a 25-kilometer grid in 6-hour increments.

Figure 1: The end-to-end WeatherNext 3 system architecture.

The model ingests live 1-hour geostationary satellite mosaics alongside traditional historical analysis to feed a single, flexible Functional Generative Network (FGN) mesh transformer.

It outputs dense gridded fields, discrete cyclone tracks, and predicts station-level sparse coordinates natively.

Figure 2: Comparison of 2-meter temperature forecasts over the UK.
WeatherNext 2 (left) at 25-kilometer (0.25°) resolution vs.
WeatherNext 3 (right) at a native 5-kilometer (0.05°) resolution.
WeatherNext 3 resolves the intricate local topography, preventing the pixelated, over-smoothed thermal representations seen in older models.

Real-world data at continuous global scale

WeatherNext 3's biggest leap forward is what it learns from.
Most AI weather models, including WeatherNext 2, are trained on data from numerical weather prediction (NWP) models.
Although useful, NWP models are complex, supercomputer-driven physics simulations that carry a six-hour data lag.
This lag can lead to biases for fast-changing variables like rain or surface temperature.

By ingesting a mosaic of live, global geostationary satellite data, our new model gains a rich, continuously updating view of the atmosphere.
This allows the model to generate a new forecast every hour, each one grounded in the most recent satellite observations available, at up to 5-kilometer resolution.

This is important because critical weather develops fast.
When storms, fronts, or precipitation systems materialize suddenly, our rapid update cycle and higher resolution provides earlier, more detailed insights needed to help drive an effective response.

Some variables, like temperature and humidity, can fluctuate dramatically over just a few kilometers, which is particularly relevant for communities near coastlines, valleys, or mountain ranges.
Traditional models struggle here because they train on representations of the atmosphere that lack detail and miss extreme local variations.

To address this, WeatherNext 3 instead trains directly on sparse weather station observation data.
This allows us to make global forecasts on a 5-kilometer grid that account for regional details like topography.

This breakthrough is particularly vital for regions across Latin America, Africa, and Asia-Pacific that have historically been underserved by high-resolution forecasting due to the immense supercomputing costs of traditional regional models.
It brings localized, high-fidelity forecasting to billions of people and local businesses in these areas.

Beyond improved resolution and forecast frequency, our model introduces predictions specifically engineered for renewable energy production.
The model forecasts 100-meter wind speeds (roughly at turbine-height) for precise wind-energy output, alongside high-resolution cloud cover and sun radiation levels to help solar farms estimate how much light they will receive on the ground.

This data is crucial for global clean energy planning, allowing grid operators and renewables developers to accurately predict how much power their clean energy assets will generate and match it with consumer demand.

Precipitation forecasting at breakthrough accuracy

Global weather models notoriously struggle to accurately predict precipitation.
Rain and snow systems are driven by fast-moving cloud processes on tiny scales that are hard to model accurately using traditional physics-based simulations.
Consequently, AI forecasts often produce blurry estimates or miss the boundaries of severe storms entirely.

To solve this, we train our model on two exceptionally high-quality sources of precipitation data: NASA’s satellite-based Integrated Multi-satellite Retrievals for GPM (IMERG) and our own global precipitation reanalysis based on satellite radar.

The result is a significant leap in precipitation forecasting accuracy.
In medium-range global forecasts, evaluations against baselines show a Continuous Ranked Probability Score (CRPS) improvement of up to 60% against IMERG, 30% for MRMS, and 10% against rain gauge measurements for early lead times.

Figure 3: Medium-range probability of precipitation (PoP > 1mm) forecast comparison.
WeatherNext 2 (left) at 25-kilometer resolution shows a highly diffused and pixelated precipitation footprint.
WeatherNext 3 (middle) at 11-kilometer resolution closely mirrors the actual satellite ground truth (right), accurately capturing the sharp, convective bands of the weather systems.

Research applied across the ecosystem

Our primary goal is to advance weather intelligence to make it universally useful — whether for an emergency responder tracking sudden wind shifts, an air traffic controller planning flight paths, or a farmer managing crops.

To bring these breakthroughs out of the lab and into the real world, we’re integrating WeatherNext 3 across Google’s core ecosystem and beyond:High-resolution forecast data: We’re making global weather predictions, updated hourly and ready to integrate into your workflows with no model setup required.
This enables researchers, developers and businesses to query the data in BigQuery and Earth Engine, or bulk-download from Google Cloud Storage.
Available globally: WeatherNext 3 will begin powering weather experiences within Google Search, Gemini app, Google Maps, Google Maps Platform Weather API, and Google Earth Engine starting today.
The update dramatically improves longer term forecasts.
When planning a day or more ahead, people will see up to 50% more accurate precipitation forecasts — with the greatest improvements in regions where forecasts have historically been less reliable.
So if you’re packing for a weekend trip or deciding the best day for an outdoor activity, you’ll now get more accurate predictions to help you plan.

The atmosphere will always retain a degree of unpredictability.
However, by training on real-world observations and bypassing traditional modeling constraints, WeatherNext 3 brings us closer to a future where forecasts truly match what is happening on the ground.

To learn more about geospatial platforms and AI work at Google, check out Google Earth Engine, AlphaEarth Foundations, and Earth AI.

Disclaimer: For official weather forecasts, severe weather warnings, and public safety advisories, please refer to your local meteorological agency or national weather service.
Learn more about WeatherNext 3 Read our paper
Build with WeatherNext 3

 

 Explore Weather Lab to see WeatherNext 3 visualized in real-time

See where WeatherNext 3 ranks on independent live leaderboards from Brightband.

 

Links : 

• Google : WeatherNext 3: Increasing resolution and WeatherNext 3: Increasing resolution and performance of global weather models withraw observations / WeatherNext: AI model achieves breakthrough in forecasting cyclones
• Bloomberg : DeepMind's new AI weather model goes hourly for power markets
• NewsCord : Google DeepMind Unveils WeatherNext 3 With Hourly, Higher-Resolution Satellite Forecasts

Inside Earth’s most remote island: ‘Almost untouched by humans’

Tristan da Cunha sits thousands of kilometers from anywhere.
Inside the tiny community that shares every job, lives with penguins and whales, and rarely sees visitors

Tristan da Cunha, the world’s most remote inhabited island.

Getty Images

From Surfer by Dashel Pierson

Tristan da Cunha is considered the world’s most remote inhabited island.
National Geographic: Tristan da Cunha is 2,700 km from South Africa, 3,700 km from South America.
Isolation shapes daily life; residents share tasks due to limited population and resources.

Looking at a map, just about halfway between the vast South Atlantic Ocean between the African and South American continents, there lies a remote, lonely island.

Actually, there’s a couple.
They’re part of a British overseas territory, but only one of them is inhabited by a small group of people, Tristan da Cunha, and for a long time it’s been dubbed the world’s most remote inhabited island.
They even have a sign.

However, below, this video explores perhaps another contender for the random title of remoteness – the Pitcairn Islands in the South Pacific.

So, which is it?

While the armchair geographer above posits that the Pitcairn Islands are the world’s most remote, many in the greater geography world still give Tristan da Cunha the title.


According to National Geographic:

“Tristan da Cunha, the most remote inhabited island in the world, is over 2,700 kilometers from South Africa and 3,700 kilometers from the nearest shores of South America.
Sitting between the South Atlantic Current to the north and the Antarctic Circumpolar Current to the south, the volcanic island and its archipelago are a hotspot of endemic biodiversity both on land and at sea.

“Among the wildlife found here are seven-gill sharks, blue sharks, shortfin mako sharks, southern right whales, fin whales, humpback whales, sperm whales, dolphins, elephant seals, and albatrosses, as well as 200,000 rockhopper penguins, more than five million shearwaters, and 300,000 sub-Antarctic fur seals.” 

In addition to the small human community, Tristan da Cunha is also home to roughly 90% of the world’s endangered Norther Rockhopper penguin population.
This species, most likely, were the inspiration for the penguins in the animate movie, Surf’s Up.

See the uncanny resemblance below. 

Rockhopper penguin (Eudyptes chrysocome moseleyi) Gough Island, South Atlantic, Islands in the southern oceans
Getty Images

So, what’s life like on the world’s most remote uninhabited island? 

 

Per NPR:

“Extreme isolation has shaped every part of life on Tristan.
With no airport and only a handful of ships visiting every year, residents say they rely largely on themselves — and each other — to keep life on the island running.

“With so few residents, there are simply too few people for all the jobs that need doing.
When someone is off island or unwell, others have to fill in, whether that means covering shifts, running errands or slaughtering a cow.
The limited labor pool means skills are shared and tasks are stretched across families, making daily life a constant balancing act.”

If you ever need to disappear, and start over in life, now you know where to go.

 

Links :

• GeoGarage blog : Extremities of the Earth: The most remote inhabited island / Vendee Globe : Tristan de Cunha passage /  New Atlantic marine sanctuary will be one of world's largest

A trio of tropical cyclones in the Pacific

IMAGE OF THE DAY FOR SEPTEMBER 4, 2026

From NASA Earth Observatory

The Pacific was buzzing with storms while the Atlantic was notably quiet.

When hurricane forecasters released their seasonal outlooks in spring 2026, the El Niñobrewing in the Pacific contributed to predictions of below-normal activity in the Atlantic basinbut above-normal activity in the northeastern and central Pacific basins.
In early September, near the climatological peak of hurricane season, those spring outlooks were on target, with the eastern Pacific buzzing with activity and the Atlantic notably quiet.

As of September 3, the Northeast Pacific had produced 15 named storms and six hurricanes, well above the norm for that point in the season.
The Atlantic basin, meanwhile, laboring under unfavorable wind shear conditions, had produced just five named storms and no hurricanes.
El Niño typically enhances hurricane activity in the eastern and central Pacific basins because of the unusually warm water temperatures it brings to those parts of the ocean.
It tends to suppress hurricane activity in the Atlantic basin by shifting large-scale circulation patterns in a way that makes it harder to sustain storms there.

At 1:14 p.m. Pacific Daylight Time (20:14 Universal Time) on September 1, NASA’s EPIC (Earth Polychromatic Imaging Camera) on the DSCOVR (Deep Space Climate Observatory) satellite captured an image of three tropical cyclones churning simultaneously in the Pacific, along with one in the Atlantic.
A band of clouds and thunderstorms associated with the Intertropical Convergence Zone (ITCZ) is visible to the south of the storms.
The spacecraft was nearly 1 million miles from Earth and just shy of 93 million miles from the Sun when the image was acquired.

 

 Visualization with NavimetriX

 

The trio of storms in the Pacific were Lowell, Karina, and Marie.
Of the three, Lowell became the strongest, with winds reaching category 5 strength for several hours on September 2.
Around the same time, Karina, spinning a few thousand kilometers to the east, achieved category 4 strength, a rare case of category 4 and 5 hurricanes occurring simultaneously in the area.
Marie, spinning southwest of Baja California, was still a tropical storm when the image was acquired but was strengthening as it moved northwest.

In the Atlantic, Tropical Storm Edouard was visible to EPIC over Louisiana and Texas, shortly after the short-lived storm made landfall.
It brought torrential rains and strong winds that downed trees and power lines.
Some areas received 15 to 24 inches (38 to 61 centimeters) of rain, according to National Weather Service meteorologists.

As of September 3, the Atlantic basin's total accumulated cyclone energy (ACE) index was 4.4, about 9 percent of normal for that date, according to statistics compiled by Colorado State University meteorologists.
Meanwhile, the Northeast Pacific basin's ACE was 130, about 50 percent above normal.
The ACE index incorporates both the intensity and longevity of storms, making it easier to compare individual storms and seasons.

Several NASA Earth-observing platforms provide data that can aid in emergency preparedness before landfall and damage assessment and response afterward.
Use the “Events” tab on NASA's Worldview browser to track current hurricanes and explore related NASA data products.

Autonomous sonar survey locates wreckage of Pan Am flight that prompted airline safety reforms

Autonomous sonar survey locates wreckage of Pan Am flight that prompted airline safety reforms

Part of the plane’s logo is still visible. 

Air/Sea Heritage Foundation and Deep Sea Vision

 

From Hydro

 

Nearly 75 years after it crashed, the wreckage of a Pan American Airways aircraft has been located and surveyed nearly 2,000 feet (610m) below the Atlantic Ocean off the coast of Puerto Rico.

A Kongsberg HUGIN autonomous underwater vehicle, equipped with advanced sonar, detected the aircraft's fuselage and tail during the first pass of the search area, with a follow-up photographic survey later confirming the identification.

The expedition was led by the Air/Sea Heritage Foundation and Deep Sea Vision, a subsidiary of Eco Minerals Inc., in partnership with the Discovery Channel's Expedition Unknownand several other organizations.

The Clipper Endeavor, a Pan Am DC-4, went down on 11 April 1952 following multiple-engine failure shortly after takeoff, forcing the pilot into a violent water landing.

The aircraft was carrying 64 passengers and five crew bound for New York.

Everyone survived the initial impact, but as the plane rapidly sank, passengers struggled to locate life vests and rafts.

Of the 69 people on board, only 17 survived.

The confusion that followed became a catalyst for new regulations requiring better flotation equipment on aircraft and pre-flight briefings on how to use it.

 

Map drawn by air force pilots

The breakthrough came after researchers uncovered records from a public hearing in Puerto Rico, including a map drawn by air force pilots who had witnessed the crash.

Combined with historical weather data from the time, this narrowed the search area to roughly 10 square nautical miles of seabed.

An earlier attempt to locate the wreckage in 2024 was abandoned due to poor weather, but the team was determined to return before the crash's 75th anniversary.

The wreckage, found broken into two sections, was located during the search area's first pass on 2 June. The subsequent photographic survey revealed that the Pan Am logo and the aircraft's name were still visible on the fuselage, providing remarkably intact confirmation of the find.

"We are all stunned and elated by this discovery, yet also humbled to remember what happened in that place so long ago," said Russ Matthews, president of the Air/Sea Heritage Foundation.

Imagery of the Clipper Endeavour wreckage, discovered more than seven decades after the aircraft disappeared. (Image courtesy: Air/Sea Heritage Foundation and Deep Sea Vision)

 

Links :

• BBC : Wreckage of Pan Am plane that shaped aviation safety found 74 years on
• Smithsonian : This 1950s Airplane Disaster Prompted the Flight Safety Demonstrations Passengers Know Today. Searchers Just Found the Wreckage in the Atlantic Ocean
• The Guardian : The 74-year mystery of Flight 526A – and six other unsolved air disasters 
• YouTube : Pan Am Plane Lost for 74 Years Found on Atlantic Ocean Floor

America’s lack of shipbuilding prowess is a problem for its navy

 

All hands on deck

photograph: getty images

 

From The Economist 

China’s commercial shipyards provide it with a big advantage

“The floating bulwark of our island” is how William Blackstone, an 18th-century British politician, described the Royal Navy.
These days it is America, not Britannia, that rules the waves.
Yet the country’s naval dominance is increasingly under threat, as China has built a floating bulwark of its own.

China’s efforts to control what it regards as its territorial waters (despite the objections of its neighbours) and project power globally have demanded a fast-expanding navy.
This has grown in short order from a small coastal-defence force to become the largest navy in the world, according to a report in 2020 by what was at the time America’s Department of Defence.
China’s 350 “battle force” vessels—including battleships, aircraft carriers, minesweepers and auxiliary craft—outnumbered America’s 293, a figure that had barely increased in two decades.

China’s plans to expand its armada to 435 ships by 2030 seem plausible.
America’s hopes of adding 58 to its fleet by 2031—and the construction of a “Trump Class” nuclear-powered battleship, proposed by the president last year—are far-fetched.
That is because China has something America sorely lacks: its rise as a naval power has been underpinned by a vast commercial-shipbuilding industry.

After the second world war Europe’s world-leading shipyards were eclipsed first by Japan, using cheap steel and labour along with new manufacturing methods, then by South Korea.
More recently it is China that has come to dominate.
Over the past 25 years its share of global shipbuilding tonnage has risen from 5% to over 50%.
America’s commercial-shipbuilding industry, by contrast, barely registers, hampering the ability of its navy to keep pace.

Matthew Funaiole of the Centre for Strategic and International Studies, a think-tank in Washington, explains that China’s commercial-shipbuilding industry is “closely intertwined” with its navy.
The fusion of military and civilian activities keeps Chinese shipyards active.
If commercial orders slow, for example, dry docks can accommodate naval work, improving the return on investment.
And although warships are far more complex than, say, container ships, the two share much in common, from steel structures and pipework to engines and propellers.
The underlying “platform” across vessels has many commonalities, points out Marzio Forlini of Bain, a consultancy.

That is why Europe, which maintains a vibrant commercial industry for specialist vessels—such as cruise ships, icebreakers and support craft for offshore energy—still has a successful naval sector that not only supplies its own forces but exports around the world.
Civilian shipyards provide more than just capacity, notes Michael Potter of Accenture, another consultancy.
They are where the essential skills needed to incorporate weapons systems, radars and other bits of defence kit are honed.
Welding, pipe-fitting and cable-pulling a commercial vessel prepares labour forces to work on complex naval ships as well.
Pierroberto Folgiero, boss of Fincantieri, Europe’s biggest shipbuilder, agrees that the two industries are highly complementary, pointing to manufacturing skills, the availability of shipyards and overlapping supply chains.
Mr Folgiero notes that the cruise ships which are Fincantieri’s speciality are hugely complicated, requiring amenities such as power plants and water systems to support 10,000 people (and 20 restaurants).
The Italian argues that if governments want naval shipbuilding, “you have to cultivate—you have to protect—civilian shipbuilding”.

In America, however, past efforts to do so have backfired.
The Jones Act, a measure introduced in 1920 to propel the domestic shipbuilding industry, has instead acted as an anchor.
It obliges transport between domestic ports to be conducted on American-built vessels (with American crews).
The result has been insufficient competition and spiralling prices: vessels manufactured in America can cost many times a similar foreign-made one.
The Jones Act—which has been temporarily suspended to allow foreign tankers to help transport oil in a bid to lower petrol prices in America—is a big part of the reason why in 2025 the country accounted for only 0.03% of global tonnage.
America’s lack of commercial-shipbuilding prowess has proved costly for its navy, whose shipyards have been unable to deliver vessels on time and on budget.
Two aircraft carriers under construction by a subsidiary of Huntington Ingalls Industries, the country’s biggest military shipbuilder, were scheduled for delivery by March 2028 but will now be over two years late.
Four-fifths of all programmes to deliver frigates, submarines and other naval vessels are behind schedule.

Politicians are aware of the problem.
The bipartisan ships for America Act proposed in 2025 aims to increase the civil fleet by 250 vessels over ten years, using subsidies and various other measures.
In February the government launched the Maritime Action Plan, which includes the creation of a $20bn fund for investments in shipbuilding.
It has also rolled out an initiative (creatively titled “Make American Shipbuilding Great Again”) to attract skills and investment from South Korea.
That effort shows some promise: last month South Korea’s three biggest shipbuilders announced 15 co-operative projects to upgrade or build new shipyards in America.
Some lawmakers have even proposed changing America’s procurement rules to allow naval vessels to be constructed by allies such as Japan or South Korea.
Most observers, however, agree there is no short-term fix to the problem.
Indeed, many are sceptical that America’s commercial-shipbuilding sector will ever again provide much support to its navy.
That ship may have sailed.

 

Links :

• CATO Institute : Washington’s Misplaced Shipbuilding Obsession
• The Atlantic : How America lost control of the Seas  
• NYTimes : America can't make what the military needs 
• The Economist : Only Asia can help America counter China’s shipbuilding prowess