Climate In A Bottle

Nvidia has a new weather forecasting product that is built to a large extent on accuracy. The Wall Street Journal calls it “climate in a bottle.

The newspaper reports, “Nvidia has unveiled a new generative foundation model that it says enables simulations of Earth’s global climate with an unprecedented level of resolution.” The product’s advantage is the ability to capture high-resolution data down to an area the size of four square miles.

Four square miles of data make forecasts more accurate. Local accuracy lends itself to accuracy across a broader area. “And to the extent that high-resolution near-term forecasts are more accurate, the accuracy of longer-term climate forecasts will improve in turn, because the accuracy of such predictions compounds over time.”

Nvidia is not alone. Google has built a product called GraphCast. The company claims its product is superior to current ones. The search company says, “GraphCast takes a significant step forward in AI for weather prediction, offering more accurate and efficient forecasts, and opening paths to support decision-making critical to the needs of our industries and societies.”

Which Model Works Best

Microsoft has a product it calls Aurora. “Over the course of Aurora’s development, researchers have been able to finetune the model to a variety of forecasting capabilities, including prediction of ocean waves and tropical cyclones, demonstrating its capability as a foundation model for the Earth system rather than just a foundation model for the atmosphere.”

All of this is to say that the current methods of weather prediction, which are decades old, are obsolete. For example, there is more than one method to track hurricanes. Under the best of circumstances, there should be one, and it should be the most accurate. AI may handle that.