Explore the vast range of titles available.

Weather forecasting is important for science and society. At present, the most accurate forecast system is the numerical weather prediction (NWP) method, which represents atmospheric states as discretized grids and numerically solves partial differential equations that describe the transition between those states 1 . However, this procedure is computationally expensive. Recently, artificial-intelligence-based methods 2 have shown potential in accelerating weather forecasting by orders of magnitude, but the forecast accuracy is still significantly lower than that of NWP methods. Here we introduce an artificial-intelligence-based method for accurate, medium-range global weather forecasting. We show that three-dimensional deep networks equipped with Earth-specific priors are effective at dealing with complex patterns in weather data, and that a hierarchical temporal aggregation strategy reduces accumulation errors in medium-range forecasting. Trained on 39 years of global data, our program, Pangu-Weather, obtains stronger deterministic forecast results on reanalysis data in all tested variables when compared with the world’s best NWP system, the operational integrated forecasting system of the European Centre for Medium-Range Weather Forecasts (ECMWF) 3 . Our method also works well with extreme weather forecasts and ensemble forecasts. When initialized with reanalysis data, the accuracy of tracking tropical cyclones is also higher than that of ECMWF-HRES. Three-dimensional deep neural networks can be trained to forecast global weather patterns, including extreme weather, with accuracy greater than or equal to that of the best numerical weather prediction models.

Weather forecasts are fundamentally uncertain, so predicting the range of probable weather scenarios is crucial for important decisions, from warning the public about hazardous weather to planning renewable energy use. Traditionally, weather forecasts have been based on numerical weather prediction (NWP) 1 , which relies on physics-based simulations of the atmosphere. Recent advances in machine learning (ML)-based weather prediction (MLWP) have produced ML-based models with less forecast error than single NWP simulations 2 , 3 . However, these advances have focused primarily on single, deterministic forecasts that fail to represent uncertainty and estimate risk. Overall, MLWP has remained less accurate and reliable than state-of-the-art NWP ensemble forecasts. Here we introduce GenCast, a probabilistic weather model with greater skill and speed than the top operational medium-range weather forecast in the world, ENS, the ensemble forecast of the European Centre for Medium-Range Weather Forecasts 4 . GenCast is an ML weather prediction method, trained on decades of reanalysis data. GenCast generates an ensemble of stochastic 15-day global forecasts, at 12-h steps and 0.25° latitude–longitude resolution, for more than 80 surface and atmospheric variables, in 8 min. It has greater skill than ENS on 97.2% of 1,320 targets we evaluated and better predicts extreme weather, tropical cyclone tracks and wind power production. This work helps open the next chapter in operational weather forecasting, in which crucial weather-dependent decisions are made more accurately and efficiently. GenCast, a probabilistic weather model using artificial intelligence for weather forecasting, has greater skill and speed than the top operational medium-range weather forecast in the world and provides probabilistic, rather than deterministic, forecasts.

Persistent heat extremes can have severe impacts on ecosystems and societies, including excess mortality, wildfires, and harvest failures. Here we identify Europe as a heatwave hotspot, exhibiting upward trends that are three-to-four times faster compared to the rest of the northern midlatitudes over the past 42 years. This accelerated trend is linked to atmospheric dynamical changes via an increase in the frequency and persistence of double jet stream states over Eurasia. We find that double jet occurrences are particularly important for western European heatwaves, explaining up to 35% of temperature variability. The upward trend in the persistence of double jet events explains almost all of the accelerated heatwave trend in western Europe, and about 30% of it over the extended European region. Those findings provide evidence that in addition to thermodynamical drivers, atmospheric dynamical changes have contributed to the increased rate of European heatwaves, with implications for risk management and potential adaptation strategies. Europe is a heatwave hotspot exhibiting three-to-four times faster upward trends compared to the rest of the northern midlatitudes. Here, this accelerated trend is linked to the increased persistence of Eurasian double jets in the upper troposphere.

Dryness stress can limit vegetation growth and is often characterized by low soil moisture (SM) and high atmospheric water demand (vapor pressure deficit, VPD). However, the relative role of SM and VPD in limiting ecosystem production remains debated and is difficult to disentangle, as SM and VPD are coupled through land-atmosphere interactions, hindering the ability to predict ecosystem responses to dryness. Here, we combine satellite observations of solar-induced fluorescence with estimates of SM and VPD and show that SM is the dominant driver of dryness stress on ecosystem production across more than 70% of vegetated land areas with valid data. Moreover, after accounting for SM-VPD coupling, VPD effects on ecosystem production are much smaller across large areas. We also find that SM stress is strongest in semi-arid ecosystems. Our results clarify a longstanding question and open new avenues for improving models to allow a better management of drought risk. Dryness stresses vegetation and can lead to declines in productivity, increased emission of carbon, and plant mortality, but the drivers of this stress remain unclear. Here the authors show that soil moisture plays a dominant role relative to atmospheric water demand over most global land vegetated areas.

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 1 . Recent advances in artificial intelligence (AI) have shown promise in improving both predictive performance and efficiency 2 , 3 , 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. Aurora, a new large-scale foundation model trained on more than one million hours of diverse geophysical data, outperforms operational forecasts in predicting air quality, ocean wave dynamics, tropical cyclone tracks and high-resolution weather.

Surface-water availability, defined as precipitation minus evapotranspiration, can be affected by changes in vegetation. These impacts can be local, due to the modification of evapotranspiration and precipitation, or non-local, due to changes in atmospheric moisture transport. However, the teleconnections of vegetation changes on water availability in downwind regions remain poorly constrained by observations. By linking measurements of local precipitation to a new hydrologically weighted leaf area index that accounts for both local and upwind vegetation contributions, we demonstrate that vegetation changes have increased global water availability at a rate of 0.26 mm yr −2 for the 2001–2018 period. Critically, this increase has attenuated about 15% of the recently observed decline in global water availability. The water availability increase is due to a greater rise in precipitation relative to evapotranspiration for over 53% of the global land surface. We also quantify the potential hydrological impacts of regional vegetation increases at any given location across global land areas. We find that enhanced vegetation is beneficial to both local and downwind water availability for ~45% of the land surface, whereas it is adverse elsewhere, primarily in water-limited or high-elevation regions. Our results highlight the potential strong effects of deliberate vegetation changes, such as afforestation programmes, on water resources beyond local and regional scales. Vegetation change over the past two decades has limited the decline in global water availability by enhancing rainfall over evapotranspiration, according to analysis of observation-based atmospheric moisture transport data.

General circulation models (GCMs) are the foundation of weather and climate prediction 1 , 2 . GCMs are physics-based simulators that combine a numerical solver for large-scale dynamics with tuned representations for small-scale processes such as cloud formation. Recently, machine-learning models trained on reanalysis data have achieved comparable or better skill than GCMs for deterministic weather forecasting 3 , 4 . However, these models have not demonstrated improved ensemble forecasts, or shown sufficient stability for long-term weather and climate simulations. Here we present a GCM that combines a differentiable solver for atmospheric dynamics with machine-learning components and show that it can generate forecasts of deterministic weather, ensemble weather and climate on par with the best machine-learning and physics-based methods. NeuralGCM is competitive with machine-learning models for one- to ten-day forecasts, and with the European Centre for Medium-Range Weather Forecasts ensemble prediction for one- to fifteen-day forecasts. With prescribed sea surface temperature, NeuralGCM can accurately track climate metrics for multiple decades, and climate forecasts with 140-kilometre resolution show emergent phenomena such as realistic frequency and trajectories of tropical cyclones. For both weather and climate, our approach offers orders of magnitude computational savings over conventional GCMs, although our model does not extrapolate to substantially different future climates. Our results show that end-to-end deep learning is compatible with tasks performed by conventional GCMs and can enhance the large-scale physical simulations that are essential for understanding and predicting the Earth system. A hybrid model that combines a differentiable solver for atmospheric dynamics with machine-learning components is capable of weather forecasts and climate simulations on par with the best machine-learning and physics-based methods.

Warming in the Arctic has been much faster than the rest of the world in both observations and model simulations, a phenomenon known as the Arctic amplification (AA) whose cause is still under debate. By analyzing data and model simulations, here we show that large AA occurs only from October to April and only over areas with significant sea-ice loss. AA largely disappears when Arctic sea ice is fixed or melts away. Periods with larger AA are associated with larger sea-ice loss, and models with bigger sea-ice loss produce larger AA. Increased outgoing longwave radiation and heat fluxes from the newly opened waters cause AA, whereas all other processes can only indirectly contribute to AA by melting sea-ice. We conclude that sea-ice loss is necessary for the existence of large AA and that models need to simulate Arctic sea ice realistically in order to correctly simulate Arctic warming under increasing CO 2 . The cause of Arctic amplification is still heavily debated. Here the authors present climate change simulations to show that sea-ice loss is essential for the existence of Arctic amplification.

Skillful subseasonal forecasts are crucial for various sectors of society but pose a grand scientific challenge. Recently, machine learning-based weather forecasting models outperform the most successful numerical weather predictions generated by the European Centre for Medium-Range Weather Forecasts (ECMWF), but have not yet surpassed conventional models at subseasonal timescales. This paper introduces FuXi Subseasonal-to-Seasonal (FuXi-S2S), a machine learning model that provides global daily mean forecasts up to 42 days, encompassing five upper-air atmospheric variables at 13 pressure levels and 11 surface variables. FuXi-S2S, trained on 72 years of daily statistics from ECMWF ERA5 reanalysis data, outperforms the ECMWF’s state-of-the-art Subseasonal-to-Seasonal model in ensemble mean and ensemble forecasts for total precipitation and outgoing longwave radiation, notably enhancing global precipitation forecast. The improved performance of FuXi-S2S can be primarily attributed to its superior capability to capture forecast uncertainty and accurately predict the Madden-Julian Oscillation (MJO), extending the skillful MJO prediction from 30 days to 36 days. Moreover, FuXi-S2S not only captures realistic teleconnections associated with the MJO but also emerges as a valuable tool for discovering precursor signals, offering researchers insights and potentially establishing a new paradigm in Earth system science research. This paper introduces FuXi-S2S, a machine-learning model that outperforms conventional numerical weather prediction models at subseasonal timescales globally, extending the skillful Madden–Julian Oscillation prediction form 30 days to 36 days.

Global climate models represent small-scale processes such as convection using subgrid models known as parameterizations, and these parameterizations contribute substantially to uncertainty in climate projections. Machine learning of new parameterizations from high-resolution model output is a promising approach, but such parameterizations have been prone to issues of instability and climate drift, and their performance for different grid spacings has not yet been investigated. Here we use a random forest to learn a parameterization from coarse-grained output of a three-dimensional high-resolution idealized atmospheric model. The parameterization leads to stable simulations at coarse resolution that replicate the climate of the high-resolution simulation. Retraining for different coarse-graining factors shows the parameterization performs best at smaller horizontal grid spacings. Our results yield insights into parameterization performance across length scales, and they also demonstrate the potential for learning parameterizations from global high-resolution simulations that are now emerging. Machine learning has been used to represent small-scale processes, such as clouds, in atmospheric models but this can lead to instability in simulations of climate. Here, the authors demonstrate a use of machine learning in an atmospheric model that leads to stable simulations of climate at a range of grid spacings.