WeatherNext 3: Increasing resolution and performance of global weather models with raw observations
Stephan Rasp and colleagues at Google Research and Google DeepMind release WeatherNext 3, which trains on raw observations rather than only reanalysis and matches physics-based models on resolution.
Ask this paper
Hourly forecasts at 0.1 degree. Previous AI weather models ran at 6-hour steps and coarser grids; WeatherNext 3 matches the temporal and spatial resolution of the best physics-based global models, including solar radiation and cloud cover.
Raw observations break the analysis dependency. Earlier models are initialized and trained only on analysis data, so they inherit its biases and cannot use new observations directly. WeatherNext 3 ingests low-latency geostationary satellite data, which is what enables the hourly refresh.
It predicts observation-space targets, including satellite-derived precipitation, tropical cyclone tracks, and station observations, rather than only the traditional analysis variables.
Station modeling gives point forecasts anywhere. Conditioning on local geography yields 2m temperature and dewpoint at arbitrary locations with substantially lower error than competing global models.
New state of the art for probabilistic medium-range skill, which is the headline claim and the reason this is the most consequential release in the block.
Abstract
State-of-the-art AI weather models have shown impressive medium-range forecast skill and computational efficiency, but suffer two key shortcomings: their forecasts have lower spatial and temporal resolution than the best physics-based models and they are exclusively initialized with and trained on analysis data. As a result, they cannot directly make use of observations, and any biases in the analysis are inherited by the forecast. WeatherNext 3 addresses these shortcomings and establishes a new state-of-the-art for probabilistic medium-range forecasting skill. First, WeatherNext 3 generates new forecasts every hour (rather than every 6 hours like traditional global models) by ingesting low-latency geostationary satellite data. Second, WeatherNext 3's temporal and spatial resolution are on par with physics-based global models, with hourly time steps and 0.1 degree resolution for single-level variables, including solar radiation and cloud cover. Third, WeatherNext 3 moves beyond traditional analysis variables by learning to predict satellite-derived precipitation estimates, as well as tropical cyclone and station observations. Modelling sparse station data allows WeatherNext 3 to make 2m temperature and dewpoint predictions at any location and time, conditioned on local geographical features, with substantially lower error than competing global models, even when evaluated against unseen stations. Together, WeatherNext 3's capabilities move operational AI-based weather forecasting beyond emulating the traditionally distinct stages of data assimilation, forecasting and post-processing, which helps to further push the frontier of performance and granularity for global weather prediction.