Search arXiv⌕ Search
Home/Archive/arXiv

arXiv · 2609.39379

Butterfly Effect Confirmed in Global AI Weather Models: Evidence from Tropical Cyclone Forecasting

Jeremy Cheuk-Hin Leung·Daosheng Xu·Weiye Yu·Shaojing Zhang·Xiaodong Zeng·Gaozhen Nie·Jie Feng·Jingchen Pu·Yi Li·Kaijun Ren·Qingcun Zeng·Banglin Zhang
View original source ↗Read source document ↗Export metadata ↓

Abstract

A paradox recently emerged in artificial intelligence (AI) weather prediction research. While some claim AI weather models cannot simulate atmospheric butterfly effect, this conflicts with AI models' limited predictability and advances in AI ensemble forecasting. This study demonstrates via counterexamples that the butterfly effect does exist in AI weather predictions. For Super Typhoon Khanun, AI predictions are constrained by a double-attractor system. Minor initial perturbations confined to two regions trigger state transitions between two local attractors, causing a 1006-km difference in the predicted storm position on Day 7. This behavior is consistent with numerical weather prediction models and observed in ~12% of tropical cyclones in the past 5 years. These findings verify AI's ability to capture atmospheric chaos and provide the physical basis for AI ensemble forecasting.

Explore related subjects

physics.ao-phnlin.CD

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Jeremy Cheuk-Hin Leung, Daosheng Xu, Weiye Yu, Shaojing Zhang, Xiaodong Zeng, Gaozhen Nie, Jie Feng, Jingchen Pu, Yi Li, Kaijun Ren, Qingcun Zeng, Banglin Zhang. 2026-09-30. Butterfly Effect Confirmed in Global AI Weather Models: Evidence from Tropical Cyclone Forecasting. https://arxiv.org/abs/2609.39379

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

KEEP EXPLORING

Related papers

arXiv2026

Scaling analysis for buoyant plumes over wildland fires

Tracking the structure and geometric properties of a buoyant plume in crosswind is critical for managing smoke hazards and improving disaster mitigation decision-making. Plume features, such as tilt angle, height, and curvature changes, are impacted by multiple forcing parameters, with canopy-induced turbulence patterns adding complexity. This study examines how these parameters, reduced to fewer dimensionless groups, affect the plume centerline slope, both near the surface and in the far field. Results from large-eddy simulations, conducted in both canopy and no-canopy environments, explore power-law dependencies, based on prior formulations, between the slopes and dimensionless groups describing the (1) relative strength of the buoyancy source to ambient wind forcing, (2) plume turbulence intensity relative to upstream turbulence, and (3) canopy density and height. Near-surface slopes are an order of magnitude higher in the canopy cases, although their sensitivity to group (1) is similar across both environments. In the canopy cases, the near-surface slope follows a one-fourth power-law dependence with group (2). This effect is absent in the no-canopy case, reflecting differences in the momentum-flux structure near the source. Moreover, the canopy aerodynamic effects delay the plume transition from the rise phase into the far-field, bent-over phase; however, canopy effects diminish as buoyancy source strength increases. The near-surface plume slope exhibits a more complex dependence on canopy geometry, increasing almost linearly with the canopy height but showing weaker sensitivity to canopy drag. Our findings support the development of scaling laws for plume structures across varied vegetative landscapes and inform improved predictive modeling efforts.

physics.ao-ph↗
arXiv2026

Long-term Variations of Cyclone-Influenced Exchange Flows in the Coastal Arctic in 2000-2025

Summertime synoptic weather regulates Arctic land-ocean exchange through meteorological tides during the ice-free period, when the predominantly easterly regime of the Beaufort High is punctuated by episodic Arctic cyclones, changing coastal winds to westerlies. We analyze wind-driven exchange flows in a northern Alaska lagoon using a 26-year reconstruction (2000-2025) validated by new in situ observations and FVCOM simulations. Phase-space EOF analysis of ERA5 data links deep cyclonic anomalies to 89% of Extreme Eastward Throughflow (EET) events (24 of 27). Coastal easterly occurrence frequencies show a consistent decrease (up to 38.3% at > 10 m/s) while low-to-moderate westerly occurrence frequencies increase by up to 30.9% but decrease by 40.6% at extreme westerly wind speeds (>10 m/s). Offshore ERA5 wind occurrence frequencies increased, with extreme offshore events (> 10 m/s) increasing by 82.4%. Local ASOS observations reveal a near-monotonic ~88% reduction in atmospheric forcing variability (East-West Wind Excursion variability drops from 2.98X10^4 to 0.35X10^4 km). The In-Out Flow Excursion variability declines steadily by 70% from 1.40X10^3 to <0.40X10^3 km, while the positive (outward) flow ratio variability contracts steadily from 7.4% to below 2.2%. Extreme Westward Throughflow (EWT) events were infrequent, but were concentrated in the most recent period, with five of six events occurring in 2022-2024 and one in 2013. These results indicate that the summertime coastal regime remains predominantly easterly, but exhibits progressively lower variability, while the wind distribution shifts toward more frequent low-to-moderate westerlies. This evolution of weather-forcing state may impact flushing and biogeochemical connectivity of Arctic lagoon ecosystems.

physics.ao-ph↗
arXiv2026

reGRAF: a global MPAS reforecast data set with convection-allowing refinements over the US and Europe

NVIDIA and The Weather Company (TWC) have generated a data set of reforecasts from TWC's GRAF (Global high-Resolution Atmospheric Forecasting) model, a version of the National Center for Atmospheric Research (NCAR) Model for Predictions Across Scales (\href{https://ncar.ucar.edu/what-we-offer/models/model-prediction-across-scales-mpas}{\underline{MPAS}}). GRAF is global, but the configuration for this reforecast had a mesh refinement to \textasciitilde4 km over the US, Caribbean Basin, and Europe, and 15 km elsewhere. This model was designed to run much of the computation on graphical processing units (GPUs), with this development assisted by NVIDIA. The 1836 reforecast cases (\textasciitilde5 years) were generated from ECMWF reanalyses (ERA5) for selected initial condition dates spanning more than 20 years, 2004--2024. These dates of the chosen initial conditions were mostly selected based on high-impact weather in the contiguous US (CONUS) and Caribbean. Sampling in this way, the reforecast spanned a wider range of interesting, high-impact weather scenarios than had we performed five contiguous years of once-daily reforecasts. The reforecast still provides many samples in non-precipitating regions with more ordinary weather. GRAF reforecasts were mostly run to $+$27 h lead time, assuming a 3-h spin-up followed by a full diurnal cycle. Data were saved in zarr format on the native model vertical coordinate. Most fields were archived at 15-min intervals, though several precipitation variables were saved at 5-min cadence. Data are made publicly available to all through Amazon Web Services' Open-Data Initiative at https://registry.opendata.aws/graf-reforecast.

physics.ao-ph↗