weather-forecasting

In this podcast episode, Peter Battaglia and Fry explore WeatherNext 3, an AI model for weather forecasting, discussing its applications in hurricane tracking and renewable energy grid optimization, along with probabilistic forecasting and future outlooks.

WeatherNext 3 is DeepMind's most advanced global weather AI model, capable of generating hourly forecasts with 5-10km resolution for applications like renewable energy and integration into Google products.

WeatherNext 3 is an advanced AI weather forecasting model from Google that incorporates real-time satellite data, hourly refreshes, and precise precipitation forecasting, now integrated into Google services like Search, Gemini, Maps, and Cloud.

A podcast discussion covers AI applications in physical simulation, including scaling to trillion-parameter models, high-resolution weather forecasting, and plasma behavior prediction in fusion energy.

The paper proposes BaguanHR, a framework that uses variable-wise super-resolution to synthesize high-resolution weather data from coarse-resolution sources, overcoming data limitations for ML-based forecasting and demonstrating power-law scaling effects for improved performance.

This article explains what El Niño is, how it is measured, its global impacts on weather and the economy, and discusses the potential for a super El Niño event.

DeepMind's hurricane AI model gives forecasters an extra day of warning and is being open-sourced as WeatherNext models, though researchers don't fully understand how it works.

Google DeepMind is open sourcing WeatherNext, an AI model that achieves state-of-the-art accuracy in cyclone forecasting, offering an extra day of warning and representing a decade of meteorological progress.

Google DeepMind's WeatherNext AI model, published in Nature, achieves state-of-the-art accuracy in cyclone track and intensity forecasting, providing an average of 24 extra hours of preparation time.

Google DeepMind's WeatherNext AI model achieves state-of-the-art cyclone forecasting accuracy, providing an extra day of warning, and is now open-sourced.

Microsoft Research's newsletter highlights Aurora 1.5 for open weather forecasting, Flint for AI-driven visualization, FlowDAgger for robot adaptation, and other advances in AI security and conversational memory.

This paper proposes TSSM, a triaxial state space model for global station weather forecasting that incorporates historical data aligned by period to improve long-horizon and extreme event prediction. It achieves state-of-the-art performance on the large-scale Weather-5K dataset and demonstrates strong robustness under missing observations.

Introduces Sparse-Reslim, a plug-in routing module that processes only 25% of spatial tokens through expensive transformer blocks for efficient weather prediction, achieving up to 3.18x speedup and improved forecast accuracy.

This paper presents NIVA, a multimodal foundation model trained on Earth system simulations to learn coupled atmosphere-ocean dynamics for subseasonal-to-seasonal prediction. Initial validation shows the model captures key modes of climate variability by accurately predicting major climate indices.

Otter Weather is a computationally efficient AI model for medium-range weather forecasting that outperforms numerical weather prediction baselines and frontier AI models while requiring significantly less training compute, aiming to democratize high-performance weather prediction.

This paper investigates the internal representations of the Aurora foundation weather model using PCA and layer-wise relevance propagation, finding that its latent space is organized by seasonal cycles and that the model learns meteorological coherence and vertical structure without explicit instruction.

The NSF's decommissioning of Ocean Station Papa, a key ocean monitoring network, will leave Alaskans with reduced weather forecasting capabilities and increased vulnerability for coastal communities.

The paper proposes Neural Tangent Kernel-based uncertainty quantification for deterministic deep learning weather models, achieving sharper adaptive prediction intervals during extreme events without retraining.

Introduces AdaWeather, an adaptive framework that combines multiple probabilistic weather forecasts using machine learning and mixture of experts, achieving logarithmic regret compared to the best static mixture of experts and showing empirical improvements in temperature forecasting.

Windborne Systems launched WeatherMesh 6, an AI weather forecasting model that claims to outperform the European Centre for Medium-Range Weather Forecasting (ECMWF) in accuracy and frequency, thanks to direct ingestion of sensor data from its balloons.