Global Models¶
-
Bridging Artificial Intelligence and Data Assimilation: The Data-driven Ensemble Forecasting System ClimaX-LETKF¶
While machine learning-based weather prediction (MLWP) has achieved significant advancements, research on assimilating real observations or ensemble forecasts within MLWP models remains limited. We...While machine learning-based weather prediction (MLWP) has achieved significant advancements, research on assimilating real observations or ensemble forecasts within MLWP models remains limited. We introduce ClimaX-LETKF, the first purely data-driven ML-based ensemble weather forecasting system. It operates stably over multiple years, independently of numerical weather prediction (NWP) models, by assimilating the NCEP ADP Global Upper Air and Surface Weather Observations. The system demonstrates greater stability and accuracy with relaxation to prior perturbation (RTPP) than with relaxation to prior spread (RTPS), while NWP models tend to be more stable with RTPS. RTPP replaces an analysis perturbation with a weighted blend of analysis and background perturbations, whereas RTPS simply rescales the analysis perturbation. Our experiments reveal that MLWP models are less capable of restoring the atmospheric field to its attractor than NWP models. This work provides valuable insights for enhancing MLWP ensemble forecasting systems and represents a substantial step toward their practical applications. more
-
A Diffusion-Based Framework for High-Resolution Precipitation Forecasting over CONUS¶
Accurate precipitation forecasting is essential for hydrometeorological risk management, especially for anticipating extreme rainfall that can lead to flash flooding and infrastructure damage. This...Accurate precipitation forecasting is essential for hydrometeorological risk management, especially for anticipating extreme rainfall that can lead to flash flooding and infrastructure damage. This study introduces a diffusion-based deep learning (DL) framework that systematically compares three residual prediction strategies differing only in their input sources: (1) a fully data-driven model using only past observations from the Multi-Radar Multi-Sensor (MRMS) system, (2) a corrective model using only forecasts from the High-Resolution Rapid Refresh (HRRR) numerical weather prediction system, and (3) a hybrid model integrating both MRMS and selected HRRR forecast variables. By evaluating these approaches under a unified setup, we provide a clearer understanding of how each data source contributes to predictive skill over the Continental United States (CONUS). Forecasts are produced at 1-km spatial resolution, beginning with direct 1-hour predictions and extending to 12 hours using autoregressive rollouts. Performance is evaluated using both CONUS-wide and region-specific metrics that assess overall performance and skill at extreme rainfall thresholds. Across all lead times, our DL framework consistently outperforms the HRRR baseline in pixel-wise and spatiostatistical metrics. The hybrid model performs best at the shortest lead time, while the HRRR-corrective model outperforms others at longer lead times, maintaining high skill through 12 hours. To assess reliability, we incorporate calibrated uncertainty quantification tailored to the residual learning setup. These gains, particularly at longer lead times, are critical for emergency preparedness, where modest increases in forecast horizon can improve decision-making. This work advances DL-based precipitation forecasting by enhancing predictive skill, reliability, and applicability across regions. more
-
FuXi-Nowcast: Environment-conditioned deep learning for severe convection nowcasting¶
Severe convection produces localized hazards that often require warnings before radar echoes fully reveal storm development. Convective initiation and the maintenance of intense convection remain...Severe convection produces localized hazards that often require warnings before radar echoes fully reveal storm development. Convective initiation and the maintenance of intense convection remain challenging for radar-only nowcasting because pre-convective signals may be absent from recent radar observations and strong echoes often decay rapidly in forecasts. Here we present FuXi-Nowcast, an environment-conditioned deep learning system that combines high-resolution observations with three-dimensional atmospheric forecasts to predict composite reflectivity, precipitation, wind gusts, and surface variables up to 12 h ahead. In April--July 2024 evaluations over East China, FuXi-Nowcast outperforms operational numerical, persistence and extrapolation baselines for reflectivity and precipitation. Case studies, diagnostics, and ablation experiments suggest that atmospheric moisture information and explicit preservation of strong convective signals contribute to forecasts of convective initiation and maintenance. These results show that environmental conditioning can mitigate important failure modes of radar-only nowcasting for high-impact convective weather. more
-
Forecasting Fails: Unveiling Evasion Attacks in Weather Prediction Models¶
With the increasing reliance on AI models for weather forecasting, it is imperative to evaluate their vulnerability to adversarial perturbations. This work introduces Weather Adaptive Adversarial...With the increasing reliance on AI models for weather forecasting, it is imperative to evaluate their vulnerability to adversarial perturbations. This work introduces Weather Adaptive Adversarial Perturbation Optimization (WAAPO), a novel framework for generating targeted adversarial perturbations that are both effective in manipulating forecasts and stealthy to avoid detection. WAAPO achieves this by incorporating constraints for channel sparsity, spatial localization, and smoothness, ensuring that perturbations remain physically realistic and imperceptible. Using the ERA5 dataset and FourCastNet (Pathak et al. 2022), we demonstrate WAAPO's ability to generate adversarial trajectories that align closely with predefined targets, even under constrained conditions. Our experiments highlight critical vulnerabilities in AI-driven forecasting models, where small perturbations to initial conditions can result in significant deviations in predicted weather patterns. These findings underscore the need for robust safeguards to protect against adversarial exploitation in operational forecasting systems. more
-
Robustness Test for AI Forecasting of Hurricane Florence Using FourCastNetv2 and Random Perturbations of the Initial Condition¶
Understanding the robustness of a weather forecasting model with respect to input noise or different uncertainties is important in assessing its output reliability, particularly for extreme weather...Understanding the robustness of a weather forecasting model with respect to input noise or different uncertainties is important in assessing its output reliability, particularly for extreme weather events like hurricanes. In this paper, we test sensitivity and robustness of an artificial intelligence (AI) weather forecasting model: NVIDIAs FourCastNetv2 (FCNv2). We conduct two experiments designed to assess model output under different levels of injected noise in the models initial condition. First, we perturb the initial condition of Hurricane Florence from the European Centre for Medium-Range Weather Forecasts (ECMWF) Reanalysis v5 (ERA5) dataset (September 13-16, 2018) with varying amounts of Gaussian noise and examine the impact on predicted trajectories and forecasted storm intensity. Second, we start FCNv2 with fully random initial conditions and observe how the model responds to nonsensical inputs. Our results indicate that FCNv2 accurately preserves hurricane features under low to moderate noise injection. Even under high levels of noise, the model maintains the general storm trajectory and structure, although positional accuracy begins to degrade. FCNv2 consistently underestimates storm intensity and persistence across all levels of injected noise. With full random initial conditions, the model generates smooth and cohesive forecasts after a few timesteps, implying the models tendency towards stable, smoothed outputs. Our approach is simple and portable to other data-driven AI weather forecasting models. more
-
Observation-driven correction of numerical weather prediction for marine winds¶
Accurate marine wind forecasts are essential for safe navigation, ship routing, and energy operations, yet they remain challenging because observations over the ocean are sparse, heterogeneous, and...Accurate marine wind forecasts are essential for safe navigation, ship routing, and energy operations, yet they remain challenging because observations over the ocean are sparse, heterogeneous, and temporally variable. We reformulate wind forecasting as observation-informed correction of a global numerical weather prediction (NWP) model. Rather than forecasting winds directly, we learn local correction patterns by assimilating the latest in-situ observations to adjust the Global Forecast System (GFS) output. We propose a transformer-based deep learning architecture that (i) handles irregular and time-varying observation sets through masking and set-based attention mechanisms, (ii) conditions predictions on recent observation-forecast pairs via cross-attention, and (iii) employs cyclical time embeddings and coordinate-aware location representations to enable single-pass inference at arbitrary spatial coordinates. We evaluate our model over the Atlantic Ocean using observations from the International Comprehensive Ocean-Atmosphere Data Set (ICOADS) as reference. The model reduces GFS 10-meter wind RMSE at all lead times up to 48 hours, achieving 45% improvement at 1-hour lead time and 13% improvement at 48-hour lead time. Spatial analyses reveal the most persistent improvements along coastlines and shipping routes, where observations are most abundant. The tokenized architecture naturally accommodates heterogeneous observing platforms (ships, buoys, tide gauges, and coastal stations) and produces both site-specific predictions and basin-scale gridded products in a single forward pass. These results demonstrate a practical, low-latency post-processing approach that complements NWP by learning to correct systematic forecast errors. more
-
Climatological benchmarking of AI-generated tropical cyclones¶
This study presents a comprehensive climatological benchmarking of tropical cyclones (TCs) generated by AI-based global weather prediction models. Using all TC events from the North Atlantic and...This study presents a comprehensive climatological benchmarking of tropical cyclones (TCs) generated by AI-based global weather prediction models. Using all TC events from the North Atlantic and Western Pacific basins between 2020 and 2025, we assess the ability of two AI models (Pangu-Weather and Aurora) to reproduce observed TC track density, climatology of storm characteristics, and physical consistency with TC theory. By comparing AI-simulated TCs with ERA5 reanalysis, we benchmark the distributions of intensity, size, forward speed, and evaluate the model's ability to credibly simulate extratropical transition. Results show that both Pangu and Aurora perform well in reproducing storm track density, forward speed distribution, and outer size distribution. Aurora shows an improved performance in simulating storm intensity compared to Pangu, with less bias in the distribution of minimum central pressure and maximum wind speed. However, both models overestimate the distribution of storm inner size (radius of maximum winds), especially for extreme events. AI models capture the relative frequency and temporal evolution of extratropical transition patterns with reasonable accuracy. The AI-simulated TCs are also less likely to conform to gradient wind balance compared to ERA5, indicating that the AI TCs may not be physically realistic in many cases. This benchmarking identifies systematic biases that can guide future corrections and support extended applications of AI models for TC hazard and risk assessment. Our work establishes a foundation for future studies using AI weather models in the context of TC climatological and hazard research. more
-
Concept drift of simple forecast models as a diagnostic of low-frequency, regime-dependent atmospheric reorganisation¶
Data-driven weather prediction models implicitly assume that the statistical relationship between predictors and targets is stationary. Under anthropogenic climate change, this assumption is...Data-driven weather prediction models implicitly assume that the statistical relationship between predictors and targets is stationary. Under anthropogenic climate change, this assumption is violated, yet the structure of the resulting concept drift remains poorly understood. Here we introduce concept drift of simple forecast models as a diagnostic of atmospheric reorganisation. Using ERA5 reanalysis, we quantify drift in spatially explicit linear models of daily mean sea-level pressure and 2\,m temperature. Models are trained on the 1950s and 2000s and evaluated on 2020 tp 2024; their performance difference defines a local, interpretable drift metric. By decomposing errors by frequency band, circulation regime and region, and by mapping drift globally, we show that drift is dominated by low-frequency variability and is strongly regime-dependent. Over the North Atlantic-European sector, low-frequency drift peaks in positive NAO despite a stable large-scale NAO pattern, while Western European summer temperature drift is tightly linked to changes in land-atmosphere coupling rather than mean warming alone. In winter, extreme high-pressure frequencies increase mainly in neutral and negative NAO, whereas structural drift is concentrated in positive NAO and Alpine hotspots. Benchmarking against variance-based diagnostics shows that drift aligns much more with changes in temporal persistence than with changes in volatility or extremes. These findings demonstrate that concept drift can serve as a physically meaningful diagnostic of evolving predictability, revealing aspects of atmospheric reorganisation that are invisible to standard deviation and storm-track metrics. more
-
On the Predictive Skill of Artificial Intelligence-based Weather Models for Extreme Events using Uncertainty Quantification¶
Accurate prediction of extreme weather events remains a major challenge for artificial intelligence-based weather prediction systems. While deterministic models such as FuXi, GraphCast, and SFNO have...Accurate prediction of extreme weather events remains a major challenge for artificial intelligence-based weather prediction systems. While deterministic models such as FuXi, GraphCast, and SFNO have achieved competitive forecast skill relative to numerical weather prediction, their ability to represent uncertainty and capture extremes is still limited. This study investigates how state-of-the-art deterministic artificial intelligence-based models respond to initial-condition perturbations and evaluates the resulting ensembles in forecasting extremes. Using four perturbation strategies (Gaussian, Perlin noise, Hemispheric Centered Bred Vectors, and Huge Ensembles), we generate 50 member ensembles for the August 2022 Pakistan floods and China heatwave, and complement these case studies with a global threshold-based evaluation. Ensemble skill is assessed against ERA5 and compared with IFS ENS and the AIFS ENS probabilistic model using deterministic and probabilistic metrics. Results show that simpler perturbations like Gaussian and Perlin noise produce similarly realistic ensemble spread and probabilistic skill as flow-based approaches like HCBV and HENS, narrowing but not closing the performance gap with numerical weather prediction ensembles, or native probabilistic models which retain the highest probabilistic skill across variables. Model choice is the dominant factor for ensemble performance, not perturbation method. Across variables, models capture temperature extremes more effectively than precipitation. These findings demonstrate that simple input perturbations can extend deterministic models toward probabilistic forecasting in hardware-constrained settings, supporting artificial intelligence-driven early warning systems. more
-
Bridging the Gap Between Bayesian Deep Learning and Ensemble Weather Forecasts¶
Weather forecasting is fundamentally challenged by the chaotic nature of the atmosphere, necessitating probabilistic approaches to quantify uncertainty. While traditional ensemble prediction (EPS)...Weather forecasting is fundamentally challenged by the chaotic nature of the atmosphere, necessitating probabilistic approaches to quantify uncertainty. While traditional ensemble prediction (EPS) addresses this through computationally intensive simulations, recent advances in Bayesian Deep Learning (BDL) offer a promising but often disconnected alternative. We bridge these paradigms through a unified hybrid Bayesian Deep Learning framework for ensemble weather forecasting that explicitly decomposes predictive uncertainty into epistemic and aleatoric components, learned via variational inference and a physics-informed stochastic perturbation scheme modeling flow-dependent atmospheric dynamics, respectively. We further establish a unified theoretical framework that rigorously connects BDL and EPS, providing formal theorems that decompose total predictive uncertainty into epistemic and aleatoric components under the hybrid BDL framework. We validate our framework on the large-scale 40-year ERA5 reanalysis dataset (1979-2019) with 0.25° spatial resolution. Experimental results show that our method not only improves forecast accuracy and yields better-calibrated uncertainty quantification but also achieves superior computational efficiency compared to state-of-the-art probabilistic diffusion models. We commit to making our code open-source upon acceptance of this paper. more
-
Attention-Enhanced Convolutional Autoencoder and Structured Delay Embeddings for Weather Prediction¶
Weather prediction is a quintessential problem involving the forecasting of a complex, nonlinear, and chaotic high-dimensional dynamical system. This work introduces an efficient reduced-order...Weather prediction is a quintessential problem involving the forecasting of a complex, nonlinear, and chaotic high-dimensional dynamical system. This work introduces an efficient reduced-order modeling (ROM) framework for short-range weather prediction and investigates fundamental questions in dimensionality reduction and reduced order modeling of such systems. Unlike recent AI-driven models, which require extensive computational resources, our framework prioritizes efficiency while achieving reasonable accuracy. Specifically, a ResNet-based convolutional autoencoder augmented by block attention modules is developed to reduce the dimensionality of high-dimensional weather data. Subsequently, a linear operator is learned in the time-delayed embedding of the latent space to efficiently capture the dynamics. Using the ERA5 reanalysis dataset, we demonstrate that this framework performs well in-distribution as evidenced by effectively predicting weather patterns within training data periods. We also identify important limitations in generalizing to future states, particularly in maintaining prediction accuracy beyond the training window. Our analysis reveals that weather systems exhibit strong temporal correlations that can be effectively captured through linear operations in an appropriately constructed embedding space, and that projection error rather than inference error is the main bottleneck. These findings shed light on some key challenges in reduced-order modeling of chaotic systems and point toward opportunities for hybrid approaches that combine efficient reduced-order models as baselines with more sophisticated AI architectures, particularly for applications in long-term climate modeling where computational efficiency is paramount. more
-
Power Ensemble Aggregation for Improved Extreme Event AI Prediction¶
This paper addresses the critical challenge of improving predictions of climate extreme events, specifically heat waves, using machine learning methods. Our work is framed as a classification problem...This paper addresses the critical challenge of improving predictions of climate extreme events, specifically heat waves, using machine learning methods. Our work is framed as a classification problem in which we try to predict whether surface air temperature will exceed its q-th local quantile within a specified timeframe. Our key finding is that aggregating ensemble predictions using a power mean significantly enhances the classifier's performance. By making a machine-learning based weather forecasting model generative and applying this non-linear aggregation method, we achieve better accuracy in predicting extreme heat events than with the typical mean prediction from the same model. Our power aggregation method shows promise and adaptability, as its optimal performance varies with the quantile threshold chosen, demonstrating increased effectiveness for higher extremes prediction. more
-
Beyond Resolution: Multi-Scale Weather and Climate Data for Alpine Renewable Energy in the Digital Twin Era -- First Evaluations and Recommendations¶
When Austrian hydropower production plummeted by 44% in early 2025 due to reduced snowpack, it exposed a critical vulnerability: standard meteorological and climatological datasets systematically...When Austrian hydropower production plummeted by 44% in early 2025 due to reduced snowpack, it exposed a critical vulnerability: standard meteorological and climatological datasets systematically fail in mountain regions that hold untapped renewable potential. This perspectives paper evaluates emerging solutions to the Alpine energy-climate data gap, analyzing datasets from global reanalyses (ERA5, 31 km) to kilometre-scale Digital Twins (Climate DT, Extremes DT, 4.4 km), regional reanalyses (ARA, 2.5 km), and next-generation AI weather prediction models (AIFS, 31 km). The multi-resolution assessment reveals that no single dataset excels universally: coarse reanalyses provide essential climatologies but miss valley-scale processes, while Digital Twins resolve Alpine dynamics yet remain computationally demanding. Effective energy planning therefore requires strategic dataset combinations validated against energy-relevant indices such as population-weighted extremes, wind-gust return periods, and Alpine-adjusted storm thresholds. A key frontier is sub-hourly (10-15 min) temporal resolution to match grid-operation needs. Six evidence-based recommendations outline pathways for bridging spatial and temporal scales. As renewable deployment expands globally into complex terrain, the Alpine region offers transferable perspectives for tackling identical forecasting and climate analysis challenges in mountainous regions worldwide. more
-
Improvement of a neural network convection scheme by including triggering and evaluation in present and future climates¶
In this study, we improve a neural network (NN) parameterization of deep convection in the global atmosphere model ARP-GEM. To take into account the sporadic nature of convection, we develop a NN...In this study, we improve a neural network (NN) parameterization of deep convection in the global atmosphere model ARP-GEM. To take into account the sporadic nature of convection, we develop a NN parameterization that includes a triggering mechanism that can detect whether deep convection is active or not within a grid-cell. This new data-driven parameterization outperforms the existing NN parameterization in present climate when replacing the original deep convection scheme of ARP-GEM. Online simulations with the NN parameterization run without stability issues. Then, this NN parameterization is evaluated online in a warmer climate. We confirm that using relative humidity instead of the specific total humidity as input for the NN (trained with present data) improves the performance and generalization in warmer climate. Finally, we perform the training of the NN parameterization with data from a warmer climate and this configuration get similar results when used in simulations in present or warmer climates. more
-
Do AI models predict storm impacts as accurately as physics-based models? A case study of the February 2020 storm series over the North Atlantic¶
The emergence of data-driven weather forecast models provides great promise for producing faster, computationally cheaper weather forecasts, compared to physics-based numerical models. However, while...The emergence of data-driven weather forecast models provides great promise for producing faster, computationally cheaper weather forecasts, compared to physics-based numerical models. However, while the performance of artificial intelligence (AI) models have been evaluated primarily for average conditions and single extreme weather events, less is known about their capability to capture sequences of extreme events, states that are usually accompanied by multiple hazards. The storm series in February 2020 provides a prime example to evaluate the performance of AI models for storm impacts. This event was associated with high surface impacts including intense surface wind speeds and heavy precipitation, amplified regionally due to the close succession of three extratropical storms. In this study, we compare the performance of data-driven models to physics-based models in forecasting the February 2020 storm series over the United Kingdom. We show that on weekly timescales, AI models tend to outperform the numerical model in predicting mean sea level pressure (MSLP), and, to a lesser extent, surface winds. Nevertheless, certain ensemble members within the physics-based forecast system can perform as well as, or occasionally outperform, the AI models. Moreover, weaker error correlations between atmospheric variables suggest that AI models may overlook physical constraints. This analysis helps to identify gaps and limitations in the ability of data-driven models to be used for impact warnings, and emphasizes the need to integrate such models with physics-based approaches for reliable impact forecasting. more
-
Benchmarking Regional Thermodynamic Trends in an AI emulator, ACE2, and a hybrid model, NeuralGCM¶
AI models have emerged as potential complements to physics-based models, but their skill in capturing observed regional climate trends with important societal impacts has not been explored. Here, we...AI models have emerged as potential complements to physics-based models, but their skill in capturing observed regional climate trends with important societal impacts has not been explored. Here, we benchmark satellite-era regional thermodynamic trends, including extremes, in an AI emulator (ACE2) and a hybrid model (NeuralGCM). We also compare the AI models' skill to physics-based land-atmosphere models. Both AI models show skill in capturing regional temperature trends such as Arctic Amplification. ACE2 outperforms other models in capturing vertical temperature trends in the midlatitudes. However, the AI models do not capture regional trends in heat extremes over the US Southwest. Furthermore, they do not capture drying trends in arid regions, even though they generally perform better than physics-based models. Our results also show that a data-driven AI emulator can perform comparably to, or better than, hybrid and physics-based models in capturing regional thermodynamic trends. more
-
Adaptive Spatio-Temporal Graphs with Self-Supervised Pretraining for Multi-Horizon Weather Forecasting¶
Accurate and robust weather forecasting remains a fundamental challenge due to the inherent spatio-temporal complexity of atmospheric systems. In this paper, we propose a novel self-supervised...Accurate and robust weather forecasting remains a fundamental challenge due to the inherent spatio-temporal complexity of atmospheric systems. In this paper, we propose a novel self-supervised learning framework that leverages spatio-temporal structures to improve multi-variable weather prediction. The model integrates a graph neural network (GNN) for spatial reasoning, a self-supervised pretraining scheme for representation learning, and a spatio-temporal adaptation mechanism to enhance generalization across varying forecasting horizons. Extensive experiments on both ERA5 and MERRA-2 reanalysis datasets demonstrate that our approach achieves superior performance compared to traditional numerical weather prediction (NWP) models and recent deep learning methods. Quantitative evaluations and visual analyses in Beijing and Shanghai confirm the model's capability to capture fine-grained meteorological patterns. The proposed framework provides a scalable and label-efficient solution for future data-driven weather forecasting systems. more
-
AI-boosted rare event sampling to characterize extreme weather¶
Weather extremes pose major societal risks, especially in a changing climate, but due to their rarity, they are difficult to study using limited observations or complex climate models. We introduce...Weather extremes pose major societal risks, especially in a changing climate, but due to their rarity, they are difficult to study using limited observations or complex climate models. We introduce AI+RES, a framework coupling fast AI weather forecasts with a high-fidelity physics model using a rare-event algorithm to efficiently characterize extremes. This approach enables the study of the statistics and physics of very rare events, such as once per millennium heatwaves at two orders-of-magnitude lower computational cost. AI+RES can be applied broadly across climate science and other fields concerned with rare events. more
-
SAFE: A Novel Approach to AI Weather Evaluation through Stratified Assessments of Forecasts over Earth¶
The dominant paradigm in machine learning is to assess model performance based on average loss across all samples in some test set. This amounts to averaging performance geospatially across the Earth...The dominant paradigm in machine learning is to assess model performance based on average loss across all samples in some test set. This amounts to averaging performance geospatially across the Earth in weather and climate settings, failing to account for the non-uniform distribution of human development and geography. We introduce Stratified Assessments of Forecasts over Earth (SAFE), a package for elucidating the stratified performance of a set of predictions made over Earth. SAFE integrates various data domains to stratify by different attributes associated with geospatial gridpoints: territory (usually country), global subregion, income, and landcover (land or water). This allows us to examine the performance of models for each individual stratum of the different attributes (e.g., the accuracy in every individual country). To demonstrate its importance, we utilize SAFE to benchmark a zoo of state-of-the-art AI-based weather prediction models, finding that they all exhibit disparities in forecasting skill across every attribute. We use this to seed a benchmark of model forecast fairness through stratification at different lead times for various climatic variables. By moving beyond globally-averaged metrics, we for the first time ask: where do models perform best or worst, and which models are most fair? To support further work in this direction, the SAFE package is open source and available at https://github.com/N-Masi/safe more
-
Evaluating Extreme Precipitation Forecasts: A Threshold-Weighted, Spatial Verification Approach for Comparing an AI Weather Prediction Model Against a High-Resolution NWP Model¶
Recent advances in AI-based weather prediction have led to the development of artificial intelligence weather prediction (AIWP) models with competitive forecast skill compared to traditional NWP...Recent advances in AI-based weather prediction have led to the development of artificial intelligence weather prediction (AIWP) models with competitive forecast skill compared to traditional NWP models, but with substantially reduced computational cost. There is a strong need for appropriate methods to evaluate their ability to predict extreme weather events, particularly when spatial coherence is important, and grid resolutions differ between models. We introduce a verification framework that combines spatial verification methods and proper scoring rules. Specifically, the framework extends the High-Resolution Assessment (HiRA) approach with threshold-weighted scoring rules. It enables user-oriented evaluation consistent with how forecasts may be interpreted by operational meteorologists or used in simple post-processing systems. The method supports targeted evaluation of extreme events by allowing flexible weighting of the relative importance of different decision thresholds. We demonstrate this framework by evaluating 32 months of precipitation forecasts from an AIWP model and a high-resolution NWP model. Our results show that model rankings are sensitive to the choice of neighbourhood size. Increasing the neighbourhood size has a greater impact on scores evaluating extreme-event performance for the high-resolution NWP model than for the AIWP model. At equivalent neighbourhood sizes, the high-resolution NWP model only outperformed the AIWP model in predicting extreme precipitation events at short lead times. We also demonstrate how this approach can be extended to evaluate discrimination ability in predicting heavy precipitation. We find that the high-resolution NWP model had superior discrimination ability at short lead times. more
-
Revealing the Potential of Learnable Perturbation Ensemble Forecast Model for Tropical Cyclone Prediction¶
Tropical cyclones (TCs) are highly destructive and inherently uncertain weather systems. Ensemble forecasting helps quantify these uncertainties, yet traditional systems are constrained by high...Tropical cyclones (TCs) are highly destructive and inherently uncertain weather systems. Ensemble forecasting helps quantify these uncertainties, yet traditional systems are constrained by high computational costs and limited capability to fully represent atmospheric nonlinearity. FuXi-ENS introduces a learnable perturbation scheme for ensemble generation, representing a novel AI-based forecasting paradigm. Here, we systematically compare FuXi-ENS with ECMWF-ENS using all 90 global TCs in 2018, examining their performance in TC-related physical variables, track and intensity forecasts, and the associated dynamical and thermodynamical fields. FuXi-ENS demonstrates clear advantages in predicting TC-related physical variables, and achieves more accurate track forecasts with reduced ensemble spread, though it still underestimates intensity relative to observations. Further dynamical and thermodynamical analyses reveal that FuXi-ENS better captures large-scale circulation, with moisture turbulent energy more tightly concentrated around the TC warm core, whereas ECMWF-ENS exhibits a more dispersed distribution. These findings highlight the potential of learnable perturbations to improve TC forecasting skill and provide valuable insights for advancing AI-based ensemble prediction of extreme weather events that have significant societal impacts. more
-
Hierarchical Graph Networks for Accurate Weather Forecasting via Lightweight Training¶
Climate events arise from intricate, multivariate dynamics governed by global-scale drivers, profoundly impacting food, energy, and infrastructure. Yet, accurate weather prediction remains elusive...Climate events arise from intricate, multivariate dynamics governed by global-scale drivers, profoundly impacting food, energy, and infrastructure. Yet, accurate weather prediction remains elusive due to physical processes unfolding across diverse spatio-temporal scales, which fixed-resolution methods cannot capture. Hierarchical Graph Neural Networks (HGNNs) offer a multiscale representation, but nonlinear downward mappings often erase global trends, weakening the integration of physics into forecasts. We introduce HiFlowCast and its ensemble variant HiAntFlow, HGNNs that embed physics within a multiscale prediction framework. Two innovations underpin their design: a Latent-Memory-Retention mechanism that preserves global trends during downward traversal, and a Latent-to-Physics branch that integrates PDE solution fields across diverse scales. Our Flow models cut errors by over 5% at 13-day lead times and by 5-8% under 1st and 99th quantile extremes, improving reliability for rare events. Leveraging pretrained model weights, they converge within a single epoch, reducing training cost and their carbon footprint. Such efficiency is vital as the growing scale of machine learning challenges sustainability and limits research accessibility. Code and model weights are in the supplementary materials. more
-
CSU-PCAST: A Dual-Branch Transformer Framework for medium-range ensemble Precipitation Forecasting¶
Accurate medium-range precipitation forecasting is essential for hydrometeorological risk management but remains challenging for both numerical weather prediction (NWP) systems and data-driven...Accurate medium-range precipitation forecasting is essential for hydrometeorological risk management but remains challenging for both numerical weather prediction (NWP) systems and data-driven models. We present CSU-PCAST, a deep learning-based ensemble forecasting framework for global precipitation prediction. The model is trained using ERA5 atmospheric and surface variables at 0.25° resolution with precipitation labels from NASA's IMERG dataset. CSU-PCAST uses 57 prognostic variables and static geographical fields to predict both atmospheric states and 6-h accumulated precipitation. The framework employs a Swin Transformer backbone with stochastic noise conditioning, temporal embeddings, and a dual-branch decoder for precipitation and non-precipitation variables. During inference, CSU-PCAST is initialized from operational GFS analyses and generates 30 ensemble members out to 15 days using an autoregressive strategy. Evaluation against GEFS over the full year of 2023 shows improved precipitation forecast skill at short lead times, including higher Critical Success Index (CSI) and lower RMSE during the first several forecast days. CSU-PCAST also reduces GEFS wet bias for light precipitation and dry bias at heavier precipitation thresholds. Probabilistic verification demonstrates lower CRPS, higher Brier Skill Scores at several thresholds, and improved ensemble reliability relative to GEFS, although both systems remain underdispersive. A case study of the Sanba extreme precipitation event further shows improved spatial structure and exceedance-probability guidance. These results demonstrate the potential of CSU-PCAST for short-to-medium-range ensemble precipitation forecasting while highlighting remaining challenges in extreme precipitation prediction and ensemble calibration. more
-
Learning Coupled Earth System Dynamics with GraphDOP¶
Interactions between different components of the Earth System (e.g. ocean, atmosphere, land and cryosphere) are a crucial driver of global weather patterns. Modern Numerical Weather Prediction (NWP)...Interactions between different components of the Earth System (e.g. ocean, atmosphere, land and cryosphere) are a crucial driver of global weather patterns. Modern Numerical Weather Prediction (NWP) systems typically run separate models of the different components, explicitly coupled across their interfaces to additionally model exchanges between the different components. Accurately representing these coupled interactions remains a major scientific and technical challenge of weather forecasting. GraphDOP is a graph-based machine learning model that learns to forecast weather directly from raw satellite and in-situ observations, without reliance on reanalysis products or traditional physics-based NWP models. GraphDOP simultaneously embeds information from diverse observation sources spanning the full Earth system into a shared latent space. This enables predictions that implicitly capture cross-domain interactions in a single model without the need for any explicit coupling. Here we present a selection of case studies which illustrate the capability of GraphDOP to forecast events where coupled processes play a particularly key role. These include rapid sea-ice freezing in the Arctic, mixing-induced ocean surface cooling during Hurricane Ian and the severe European heat wave of 2022. The results suggest that learning directly from Earth System observations can successfully characterise and propagate cross-component interactions, offering a promising path towards physically consistent end-to-end data-driven Earth System prediction with a single model. more
-
Signature Kernel Scoring Rule: A Spatio-Temporal Diagnostic for Probabilistic Weather Forecasting¶
Modern weather forecasting has increasingly transitioned from numerical weather prediction (NWP) to data-driven machine learning forecasting techniques. While these new models produce probabilistic...Modern weather forecasting has increasingly transitioned from numerical weather prediction (NWP) to data-driven machine learning forecasting techniques. While these new models produce probabilistic forecasts to quantify uncertainty, their training and evaluation may remain hindered by conventional scoring rules, primarily MSE, which are designed for single time point predictions and ignore the highly correlated data structures present in weather behaviour. This work introduces the signature kernel scoring rule to the domain of weather forecasting, which reframes weather variables as continuous paths to encode temporal and spatial dependencies through iterated integrals. Validated as strictly proper through the use of path augmentations to guarantee uniqueness, the signature kernel provides a theoretically robust metric for forecast verification and model training. Empirical evaluations through weather scorecards on WeatherBench 2 models demonstrate the signature kernel scoring rule's high discriminative power and unique capacity to capture path-dependent interactions. Following previous demonstration of successful adversarial-free probabilistic training, we train sliding window generative neural networks using a predictive-sequential scoring rule on ERA5 reanalysis weather data. Using a lightweight model, we demonstrate that signature kernel based training outperforms climatology for forecast paths of up to fifteen timesteps. more
-
Deploying Atmospheric and Oceanic AI Models on Chinese Hardware and Framework: Migration Strategies, Performance Optimization and Analysis¶
With the growing role of artificial intelligence in climate and weather research, efficient model training and inference are in high demand. Current models like FourCastNet and AI-GOMS depend heavily...With the growing role of artificial intelligence in climate and weather research, efficient model training and inference are in high demand. Current models like FourCastNet and AI-GOMS depend heavily on GPUs, limiting hardware independence, especially for Chinese domestic hardware and frameworks. To address this issue, we present a framework for migrating large-scale atmospheric and oceanic models from PyTorch to MindSpore and optimizing for Chinese chips, and evaluating their performance against GPUs. The framework focuses on software-hardware adaptation, memory optimization, and parallelism. Furthermore, the model's performance is evaluated across multiple metrics, including training speed, inference speed, model accuracy, and energy efficiency, with comparisons against GPU-based implementations. Experimental results demonstrate that the migration and optimization process preserves the models' original accuracy while significantly reducing system dependencies and improving operational efficiency by leveraging Chinese chips as a viable alternative for scientific computing. This work provides valuable insights and practical guidance for leveraging Chinese domestic chips and frameworks in atmospheric and oceanic AI model development, offering a pathway toward greater technological independence. more
-
Leveraging Teleconnections with Physics-Informed Graph Attention Networks for Long-Range Extreme Rainfall Forecasting in Thailand¶
Accurate rainfall forecasting, particularly for extreme events, remains a significant challenge in climatology and the Earth system. This paper presents novel physics-informed Graph Neural Networks...Accurate rainfall forecasting, particularly for extreme events, remains a significant challenge in climatology and the Earth system. This paper presents novel physics-informed Graph Neural Networks (GNNs) combined with extreme-value analysis techniques to improve gauge-station rainfall predictions across Thailand. The model leverages a graph-structured representation of gauge stations to capture complex spatiotemporal patterns, and it offers explainability through teleconnections. We preprocess relevant climate indices that potentially influence regional rainfall. The proposed Graph Attention Network with Long Short-Term Memory (Attention-LSTM) applies the attention mechanism using initial edge features derived from simple orographic-precipitation physics formulation. The embeddings are subsequently processed by LSTM layers. To address extremes, we perform Peak-Over-Threshold (POT) mapping using the novel Spatial Season-aware Generalized Pareto Distribution (GPD) method, which overcomes limitations of traditional machine-learning models. Experiments demonstrate that our method outperforms well-established baselines across most regions, including areas prone to extremes, and remains strongly competitive with the state of the art. Compared with the operational forecasting system SEAS5, our real-world application improves extreme-event prediction and offers a practical enhancement to produce high-resolution maps that support decision-making in long-term water management. more
-
Adversarial Attacks on Downstream Weather Forecasting Models: Application to Tropical Cyclone Trajectory Prediction¶
Deep learning-based weather forecasting (DLWF) models leverage past weather observations to generate future forecasts, supporting a wide range of downstream applications, including tropical cyclone...Deep learning-based weather forecasting (DLWF) models leverage past weather observations to generate future forecasts, supporting a wide range of downstream applications, including tropical cyclone (TC) prediction. In this paper, we investigate their vulnerability to adversarial attacks, where subtle perturbations to the upstream forecasts can alter the downstream TC trajectory predictions. Although research into adversarial attacks on DLWF models has grown recently, it remains challenging to craft perturbed upstream forecasts that steer the downstream outputs toward attacker-specified trajectories. First, conventional TC detection systems are opaque, non-differentiable black boxes, making standard gradient-based attacks infeasible. Second, the extreme rarity of TC events leads to severe class imbalance problem, making it difficult to develop attack methods for perturbing upstream forecasts that produce realistic-looking cyclone paths aligned with attacker's target trajectories. To overcome these limitations, we propose Cyc-Attack, a novel method for perturbing the upstream forecasts of DLWF models to generate adversarial trajectories. The proposed method uses a differentiable surrogate model to approximate the TC detector's output, enabling the application of gradient-based attacks. Cyc-Attack also employs a skewness-aware loss function with kernel dilation strategy to address the imbalance problem. Finally, a distance-based gradient weighting scheme and regularization are used to constrain the perturbations and eliminate unrealistic-looking trajectories, thereby making the adversarial upstream forecasts less easily detectable. Our experiments show that Cyc-Attack achieves a higher true positive rate in matching the attacker's target trajectories, along with lower false alarm rates and stealthier perturbations than conventional attack methods. more
-
ARROW: An Adaptive Rollout and Routing Method for Global Weather Forecasting¶
Weather forecasting is a fundamental task in spatiotemporal data analysis, with broad applications across a wide range of domains. Existing data-driven forecasting methods typically model atmospheric...Weather forecasting is a fundamental task in spatiotemporal data analysis, with broad applications across a wide range of domains. Existing data-driven forecasting methods typically model atmospheric dynamics over a fixed short time interval, e.g., 6 hours, and rely on naive autoregression-based rollout for long-term forecasting, e.g., 5 days. However, this paradigm suffers from two key limitations: (1) it often inadequately models the spatial and multi-scale temporal dependencies inherent in global weather systems, and (2) the rollout strategy struggles to balance error accumulation with the capture of fine-grained atmospheric variations. In this study, we propose ARROW, an Adaptive-Rollout Multi-scale temporal Routing method for Global Weather Forecasting. To contend with the first limitation, we construct a multi-interval forecasting model that forecasts weather across different time intervals. Within the model, the Shared-Private Mixture-of-Experts captures both shared patterns and specific characteristics of atmospheric dynamics across different time scales, while Ring Positional Encoding accurately encodes the circular latitude structure of the Earth when representing spatial information. For the second limitation, we develop an adaptive rollout scheduler based on reinforcement learning, which selects the most suitable time interval to forecast according to the current weather state. Experimental results demonstrate that ARROW achieves state-of-the-art performance in global weather forecasting, establishing a promising paradigm in this field. more
-
Control-Augmented Autoregressive Diffusion for Data Assimilation¶
Despite advances in test-time scaling and diffusion finetuning, guidance for Auto-Regressive Diffusion Models (ARDMs) remains underexplored. We introduce an amortized framework that augments a...Despite advances in test-time scaling and diffusion finetuning, guidance for Auto-Regressive Diffusion Models (ARDMs) remains underexplored. We introduce an amortized framework that augments a pretrained ARDM with an offline-trained controller. By previewing future rollouts, the controller learns stepwise corrections that anticipate observations under a terminal-cost objective, yielding a reusable policy for guided generation. Motivated by a stochastic optimal control view of ARDM trajectories, our method injects small controls within each denoising sub-step while staying close to the pretrained dynamics. We study this approach for dataassimilation (DA) in chaotic spatiotemporal partial differential equations (PDEs), where existing methods are often computationally expensive and susceptible to forecast drift under sparse observations. At inference, DA becomes a feed-forward rollout with on-the-fly corrections, achieving an order-of-magnitude speedup over strong diffusion-based baselines. Across two canonical PDEs and a compact ECMWF Reanalysis v5 (ERA5) pilot spanning six observation regimes, our method consistently improves stability and accuracy over state-of-the-art alternatives, with similar improvements observed in a larger-scale GenCast study. more