ML in high-resolution weather forecasting

Me and my colleagues at Google DeepMind and Google Research are sharing our latest work on tropical cyclone prediction, now available through a research tool, Weather Lab: https://lnkd.in/dNtjmiYq Over the past 50 years, tropical cyclones, also known as hurricanes or typhoons, have claimed more than 779,000 lives and caused $1.4 trillion in economic losses [WMO]. For the millions of people living in their path, the accuracy of weather forecasting is the most critical line of defense. In an effort to protect lives and property from this threat, we’ve built a powerful new machine learning (ML)-based ensemble weather model, deployed it operationally on Weather Lab, and partnered with experts from the U.S. National Hurricane Center (NHC) who will assess its live predictions alongside their established forecasting tools. The ensemble mean cyclone track of our new model gains about 1.5 days of position error advantage over ECMWF ENS in tests based on NHC protocols. And surprisingly, our model has a lower average intensity error than NOAA’s high-resolution hurricane model, HAFS-A, in more than 60 of the 74 cyclones evaluated in 2023 and 2024 in the East Pacific and North Atlantic basins. We achieved this by building a new kind of ML weather model, FGN [Ferran Alet Puig et al., 2025], which substantially outperforms GenCast on probabilistic metrics, and specialising it for cyclone tracking by training it on a record of nearly 5,000 tropical cyclones from the past 45 years. Most human forecasters do not trust a weather model until its performance is demonstrated in a real-time setting. That’s why we built Weather Lab, available globally, providing access to live and historical visualisations of tropical cyclone predictions from our new ML weather model, with WeatherNext and ECMWF models shown for comparison. We recently enabled live data downloads in CSV and ATCF format for experts to evaluate. This is a powerful new tool in the toolbox, but no single model is perfect. It will remain key that human forecasters evaluate a wide range of both ML and physics-based predictions when issuing public warnings for cyclone threats. And of course, ML weather models continue to depend on the historical and real-time availability of atmospheric analysis datasets produced by physical modelling centres, and the continued quality and coverage of the Earth’s observing system. Tropical cyclones will likely become more destructive over time [IPCC, 2023]. It is crucial we continue improving our monitoring, prediction, and understanding of these complex beasts of physics. Try Weather Lab: https://lnkd.in/dNtjmiYq  Blog post: https://lnkd.in/dkj8cYan  FGN (Alet et al., 2025): https://lnkd.in/dJhP9Kj2  WMO: https://lnkd.in/dPt94VX5 IPCC, 2023: https://lnkd.in/dj5n-Rqg 

Excited to announce the public code release of Aurora - a foundation model for atmospheric forecasting! 🌍 ⛅ Code: https://lnkd.in/dsaGx_hr Docs: https://lnkd.in/d7TQ-HTN Paper: https://lnkd.in/dBvgfPDG Aurora sets a new state-of-the-art in global weather and air quality prediction, outperforming traditional numerical models while being orders of magnitude faster. Key features: • Pretrained on diverse atmospheric data. • Fine-tuned versions for weather and air quality. • 0.1° resolution global forecasts. • Outperforms IFS-HRES and GraphCast on most metrics. The repo currently includes: • Pretrained model weights. • Fine-tuned weights for high-res weather forecasting. • Easy-to-use Python API. • Detailed documentation and examples. • Get started now with a simple example that runs Aurora on ERA5: https://lnkd.in/dnV5rR_V We hope this accelerates research into foundation models for Earth system prediction. Read the full paper here: https://lnkd.in/dBvgfPDG. Amazing effort by Cristian Bodnar, Wessel B., Ana Lucic and Megan Stanley at Microsoft Research AI for Science. #MachineLearning #WeatherForecasting

GenCast is out in Nature Magazine! It's the first high res ML ensemble weather forecast which outperforms the operational state of the art. And a few more things have happened since the preprint was first released ⬇️ ⬇️ ⬇️ The paper shows GenCast provided better probabilistic weather forecasts, including better forecast of extreme weather, than the operational gold standard over our year-long evaluation period. This could mean earlier preparation for extreme events, more reliable wind power, and much more. Alongside publication in Nature Magazine, we are making the code and model weights available to the community (incl. a mini version of the model which gives passable results and can run in a free colab). And soon we'll share an archive of model historical and current forecasts. We also recently fine tuned the model to work on operational inputs so that it can be run live. We conducted a retrospective analysis of this model's forecasts of the track Hurricane Milton 🌪️ . GenCast predicted 60-80% probability of landfall in Florida already from 8.5 days before landfall eventually happened - a couple of days before Milton even formed - and more than 90% from 5.75 days before. *caveats on this example* a) individual examples should always be taken with a pinch of salt - we need rigorous evaluation over an extended period, see the cyclone track section of the paper. b) these are track predictions, not intensity predictions. Overall, GenCast marks something of an inflection point in the advance of AI for weather prediction, with SOTA raw forecasts now coming from AI. I think we can expect them to be increasingly incorporated operationally alongside traditional models (and to continue to improve!) Check out the paper: https://lnkd.in/dsasGbNb The code: https://lnkd.in/dHSjfW-3 And the blog: https://shorturl.at/NPvwL Work with a remarkable team: Alvaro Sanchez Gonzalez, Ferran Alet Puig, Tom Andersson, Andrew El-Kadi, Dominic Masters, Timo Ewalds, Jacklynn Stott, Shakir Mohamed, Peter Battaglia, Rémi Lam, & Matthew Willson

🌍 Climate scientists often face a trade-off: Global Climate Models (GCMs) are essential for long-term climate projections — but they operate at coarse spatial resolution, making them too crude for regional or local decision-making. To get fine-scale data, researchers use Regional Climate Models (RCMs). These add crucial spatial detail, but come at a very high computational cost, often requiring supercomputers to run for months. ➡️ A new paper introduces EnScale — a machine learning framework that offers an efficient and accurate alternative to running full RCM simulations. Instead of solving the complex physics from scratch, EnScale "learns" the relationship between GCMs and RCMs by training on existing paired datasets. It then generates high-resolution, realistic, and diverse regional climate fields directly from GCM inputs. What makes EnScale stand out? ✅ It uses a generative ML model trained with a statistically principled loss (energy score), enabling probabilistic outputs that reflect natural variability and uncertainty ✅ It is multivariate – it learns to generate temperature, precipitation, radiation, and wind jointly, preserving spatial and cross-variable coherence ✅ It is computationally lightweight – training and inference are up to 10–20× faster than state-of-the-art generative approaches ✅ It includes an extension (EnScale-t) for generating temporally consistent time series – a must for studying events like heatwaves or prolonged droughts This approach opens the door to faster, more flexible generation of regional climate scenarios, essential for risk assessment, infrastructure planning, and climate adaptation — especially where computational resources are limited. 📄 Read the full paper: EnScale: Temporally-consistent multivariate generative downscaling via proper scoring rules ---> https://lnkd.in/dQr5rmWU (code: https://lnkd.in/dQk_Jv8g) 👏 Congrats to the authors — a strong step forward for ML-based climate modeling! #climateAI #downscaling #generativeAI #machinelearning #climatescience #EnScale #RCM #GCM #ETHZurich #climatescenarios

Aurora: Open-weight foundation model for the earth system. Trained on 1 million hours of geophysical data, to support forecasting to prepare for natural disasters. “The potential implications of Aurora for the field of Earth system prediction are profound. Although in this paper we showcase the application of Aurora to four domains, it could be fine-tuned for any desired Earth system prediction task, potentially producing fore- casts that outperform the current operational systems at a fraction of the cost. Some examples include predicting ocean circulation, local and regional weather, seasonal weather, vegetation growth and phenology, extreme weather modalities such as floods and wildfires, pollination patterns, agricultural productivity, renewable energy production and sea ice extent. With the ability to fine-tune Aurora to diverse application domains at only modest computational cost, Aurora represents notable progress in making actionable predictions accessible to anyone.” Aurora was developed by ML researchers and domain experts in meteorology and earth system modelling at Microsoft Research AI for Science in Amsterdam and Cambridge, UK. Link to paper Bodnar et al (2025) in the comments. Abstract Reliable forecasting of the Earth system is essential for mitigating natural disasters and supporting human progress. Traditional numerical models, although powerful, are extremely computationally expensive. Recent advances in artificial intelligence (AI) have shown promise in improving both predictive performance and efficiency, yet their potential remains underexplored in many Earth system domains. Here we introduce Aurora, a large-scale foundation model trained on more than one million hours of diverse geophysical data. Aurora outperforms operational forecasts in predicting air quality, ocean waves, tropical cyclone tracks and high-resolution weather, all at orders of magnitude lower computational cost. With the ability to be fine-tuned for diverse applications at modest expense, Aurora represents a notable step towards democratizing accurate and efficient Earth system predictions. These results highlight the transformative potential of AI in environmental forecasting and pave the way for broader accessibility to high-quality climate and weather information. #geoscience #earthscience #AI #artificialintelligence #geohazards #weather #climate #meteorology

AI Replaces Supercomputers in Weather Forecasting with Instant Predictions A Forecasting Revolution in One Second New research reveals that artificial intelligence can now deliver weather forecasts in seconds on a desktop—matching the accuracy of traditional models that require hours or days on supercomputers. This breakthrough marks a dramatic shift in meteorology, where the reliance on physics-based numerical weather prediction (NWP) models—unchanged in principle since the 1950s—is being replaced by AI-driven forecasting that is faster, cheaper, and far more energy-efficient. How AI is Disrupting Traditional Forecasting Historically, weather predictions depend on vast data inputs from satellites, balloons, and weather stations. These observations are fed into complex NWP models that simulate the atmosphere based on physical laws, requiring massive computational resources. • Heavy Computing Burden: Running these models demands high-performance supercomputers, consuming significant time, power, and budget. • AI as a Lightweight Alternative: The new AI model operates in a single second on desktop hardware, offering comparable forecast accuracy without the need for physics-based simulation. • Machine Learning Core: Rather than modeling physical processes, the AI learns directly from decades of historical data to detect patterns and predict atmospheric conditions. From Patchwork Improvements to Full Replacement The journey toward full AI-driven forecasting began with hybrid models. Google researchers developed AI tools that could optimize select components of traditional systems, reducing computational loads. DeepMind, Google’s AI subsidiary, went further by creating graph-based models that completely replaced the forecasting process. • European Adoption: The European Centre for Medium-Range Weather Forecasts (ECMWF) has already begun using AI-based systems, marking one of the first institutional adoptions of this new approach. • ForecastNet and GraphCast: These AI models use neural networks trained on historical weather data to predict temperature, pressure, precipitation, and wind with high spatial and temporal resolution. Why It Matters AI’s success in weather forecasting is not just a technological achievement—it’s a paradigm shift with implications for science, society, and sustainability. • Energy Efficiency: AI reduces the carbon footprint of weather forecasting by orders of magnitude—critical in an era of climate awareness. • Faster Response for Emergencies: Rapid forecasts can assist governments and aid agencies in responding to severe weather, such as hurricanes, wildfires, and floods, in near real-time. This milestone signals a profound transformation in how we understand and anticipate weather. By turning historical data into actionable predictions in seconds, AI is not just optimizing forecasts—it is rewriting the very foundations of meteorology.

Another new milestone is broken by AI. Just two years after producing a better short-term forecast model, a recent model has now comprehensively beaten the existing weather models much, much less compute power. 💥 But how its beaten it, and when, is where the interesting part begins. .. A new model from Google's Deepming ('GenCast') has beaten our best models in forecasting weather events. This is a big deal for two reasons: 1. Climate is changing, and bigger storms, and bigger economic events are happening. Accurate and fast weather predictions are crucial to mitigating the economic impact of extreme conditions. Faster, more precise forecasts can help governments, businesses, and communities act sooner—reducing the damage from floods, heatwaves, and other disruptive events. The forecast of weather, specifically winds, is also critical for our use of wind energy, and turbine management. To avoid damaging infrastructure, and optimally switch these on and off for example (Which I'm told is extremely expensive to do regularly). 2. Traditionally, weather prediction relies on numerical weather prediction (NWP) models, which simulate complex atmospheric processes using physical equations. While powerful, NWPs require enormous computational resources and often struggle to deliver high-resolution forecasts quickly enough, particularly for extreme conditions. So importantly, we are not talking about the 'middle' of the distribution, but rather around the edges, and unusual events. GenCast, by contrast, is a diffusion model that generates highly accurate forecasts at lower computational costs and with faster turnaround times. Fourty years of training, it generates many scenarios in parallel, effectively creating distributions of uncertainty. This approach enhances the prediction of severe weather events, allowing for more timely warnings and better preparation. Importantly, and again, the advent of machine learning means we are allowing data and processes to learn patterns (potentially not linked or driven by our understanding of the underlying causal processes). This should make us nervous, because again, 'it just works'. The paper: https://lnkd.in/gavxZvjX (they open sourced the model!) What are your thoughts on the role of AI in tackling climate risks? Bureau of Meteorology Andrew Huang, CFA Well done the team at: Google DeepMind #climate #esg #ai #artificialintelligence #genai Marcos Lopez de Prado

Aurora, a 1.3 billion parameter foundation model from Microsoft Research AI for Science, offers high-resolution weather and air pollution forecasts by leveraging diverse atmospheric data. It efficiently adapts to new tasks, outperforms traditional models in extreme weather prediction, and delivers rapid results, all while being computationally efficient. For municipalities, it provides rapid and accurate predictions, aiding in better decision-making for urban planning, emergency response, and environmental management, while efficiently handling extreme weather events. Paper: https://lnkd.in/dMa6Tkya Github: https://lnkd.in/dwJYA3WY Infographics: Napkin AI #AI #ClimateScience #WeatherForecasting #MunicipalPlanning #UrbanManagement #EmergencyResponse #EnvironmentalManagement #MachineLearning #SmartCities #PublicSafety