Search arXiv⌕ Search
Home/Archive/arXiv

arXiv · 2511.21792

Climatological benchmarking of AI-generated tropical cyclones

Yanmo Weng·Avantika Gori
View original source ↗Read source document ↗Export metadata ↓

Abstract

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.

Explore related subjects

physics.ao-ph

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Yanmo Weng, Avantika Gori. 2025-11-26. Climatological benchmarking of AI-generated tropical cyclones. https://arxiv.org/abs/2511.21792

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

The impact of two-dimensional filtering on SWOT observations with white noise

The Surface Water and Ocean Topography (SWOT) mission provides two-dimensional observations of sea surface height (SSH) at unprecedented spatial resolution, enabling exploration of ocean variability down to scales of $O(10~\mathrm{ km})$. At these scales, however, interpreting SSH variability is challenging because ocean dynamical signals overlap with measurement noise, and their respective spectral signatures are not yet fully understood. Recent analyses of SWOT 2-km posting observations have shown that along-track spectra transition to flatter but still red spectra at scales around 30 km. These flatter portions of the spectra have a power-law-like behavior and spectral slopes of approximately $-1$ or steeper, and their magnitudes and slopes are correlated with SWOT measurement noise magnitude. Here, we investigate the hypothesis that these flatter but still red along-track small-scale spectra can arise from two-dimensional filtering and aliasing of spatially uncorrelated (white) noise. Using synthetic experiments, we show that the resulting one-dimensional along-track spectra exhibit a similar transition to red, power-law-like behavior at scales of 15--50 km, qualitatively similar to spectral behavior reported in SWOT observations. The transition scale and apparent spectral slopes depend on the noise level, its cross-track variability, and the background ocean signal. This finding highlights the importance of carefully accounting for measurement noise and processing effects when interpreting SWOT spectra, and suggests that a white noise model should serve as a baseline null hypothesis for small-scale spectral analyses.

physics.ao-ph↗
arXiv2026

Improving global precipitation forecasts with an AI weather model trained on satellite observations

Precipitation forecasts shape decision-making across the global economy, particularly in sectors such as agriculture. However, unlike variables such as temperature, precipitation is highly intermittent and localized, making it difficult to forecast. While recent advances in AI weather prediction systems have enabled them to surpass physical models on globally averaged metrics, improvements in mean error rarely translate to actionable forecasts of severe flooding or dry crop fields. Furthermore, most of these models are trained and evaluated against a reanalysis data product, ERA5, which has well-known biases. Here we retrain AIFS, ECMWF's widely-used, open-source operational 0.25° probabilistic graph-transformer weather model, on satellite-based precipitation observations. Our model, Laxmi, improves global probabilistic accuracy by 19% and resolves systematic distributional biases in ERA5. Specifically, Laxmi reduces drizzle overprediction by 33% for amounts less than 3 mm per day. It also mitigates extreme rainfall underprediction, improving the global 95th percentile Brier skill score by 57%. Across a case study of 10 Indian tropical storms, Laxmi delivered the most accurate forecast of 150 mm event-total precipitation in 7 events, compared to 1 for AIFS and 2 for the leading physical model, IFS. Our results demonstrate that incorporating observation-based precipitation data directly into training can substantially improve forecasts.

physics.ao-ph↗
arXiv2026

HClimRep-Ocean: A Global Ocean Emulator on an Unstructured Mesh

Machine-learning (ML) emulators for atmospheric processes have advanced rapidly in recent years, transforming weather forecasting. Although early ML ocean forecasting models now exist, they remain less developed than their atmospheric counterparts. Unlike the atmosphere, much of the ocean's kinetic energy resides in mesoscale eddies whose characteristic spatial scales are approximately an order of magnitude smaller than those of comparable atmospheric features. Moreover, complex coastlines, narrow straits, and ice-covered seas make boundary representation a central challenge that atmospheric models do not face. Consequently, numerical ocean simulations commonly use locally refined or even completely unstructured meshes. However, their data-driven counterparts have so far been built around latitude-longitude grids. We present HClimRep-Ocean, an ocean emulator that operates directly on the native unstructured mesh of FESOM2. The emulator is trained on a 209-year AWI-CM3 control integration and is run without atmospheric forcing, receiving the atmospheric state only at initialisation time, which isolates the predictability carried by the ocean state itself. Skill is strongly field-dependent: for currents, HClimRep-Ocean outperforms every reference at 30 day forecast, whereas for temperature and salinity a damped-anomaly persistence forecast remains the more accurate estimator. This behaviour is physically interpretable: current variability is largely geostrophic and internally generated, whereas sea-surface temperature and salinity fluctuations are driven by atmospheric forcing through weather state. Evaluated independently on the OceanBench benchmark, a reanalysis-trained variant of HClimRep-Ocean achieves the lowest RMSE against GLORYS reanalysis among all assessed systems, confirming the competitiveness of the native-mesh approach.

physics.ao-ph↗