Enhancing AI-Based Tropical Cyclone Track and Intensity Forecasting via Systematic Bias Correction

Peisong Niu; Haifan Zhang; Yang Zhao; Tian Zhou; Ziqing Ma; Wenqiang Shen; Junping Zhao; Huiling Yuan

cs.LGMar 20, 2026

Enhancing AI-Based Tropical Cyclone Track and Intensity Forecasting via Systematic Bias Correction

Authors: Peisong Niu, Haifan Zhang, Yang Zhao, Tian Zhou, Ziqing Ma, Wenqiang Shen, Junping Zhao, Huiling Yuan,

Organizations: DAMO Academy, Alibaba Group Bellevue, USA

Abstract

Tropical cyclones (TCs) pose severe threats to life, infrastructure, and economies in tropical and subtropical regions, underscoring the critical need for accurate and timely forecasts of both track and intensity. Recent advances in AI-based weather forecasting have shown promise in improving TC track forecasts. However, these systems are typically trained on coarse-resolution reanalysis data (e.g., ERA5 at 0.25 degree), which constrains predicted TC positions to a fixed grid and introduces significant discretization errors. Moreover, intensity forecasting remains limited especially for strong TCs by the smoothing effect of coarse meteorological fields and the use of regression losses that bias predictions toward conditional means. To address these limitations, we propose BaguanCyclone, a novel, unified framework that integrates two key innovations: (1) a probabilistic center refinement module that models the continuous spatial distribution of TC centers, enabling finer track precision; and (2) a region-aware intensity forecasting module that leverages high-resolution internal representations within dynamically defined sub-grid zones around the TC core to better capture localized extremes. Evaluated on the global IBTrACS dataset across six major TC basins, our system consistently outperforms both operational numerical weather prediction (NWP) models and most AI-based baselines, delivering a substantial enhancement in forecast accuracy. Remarkably, BaguanCyclone excels in navigating meteorological complexities, consistently delivering accurate forecasts for re-intensification, sweeping arcs, twin cyclones, and meandering events. Our code is available at https://github.com/DAMO-DI-ML/Baguan-cyclone.

Figures & tables

Figure 1. Tropical cyclone intensities and tracks. IBTrACS data (2000–2024) colored by Saffir-Simpson category.

Figure 2. Overview of BaguanCyclone’s architecture. (a) The input consists of an initial position and 2 subsequent forecast frames. (b) The probabilistic center refinement model first employs a nonparametric algorithm to obtain an initial tracking distribution, which is then refined using a tracking correction model. (c) Region-aware intensity forecasting model leverages a Swin-transformer with patch merging to compute the per-region intensity.

	dist. (km)		wind (m/s)
	ERA5 ( $0.25^{\circ}$ )	EC Analysis ( $0.1^{\circ}$ )	ERA5 ( $0.25^{\circ}$ )	EC Analysis ( $0.1^{\circ}$ )
WP	45.3	56.9	7.74	5.69
NA	32.7	48.1	11.53	8.87

Table 1. Comparison of ERA5 (

0.25^{\circ}

) and EC Analysis (

0.1^{\circ}

) over the Western North Pacific (WP) and North Atlantic (NA). Results are presented only for 6-hour.

Basin	WP	EP	NA	NI	SI	SP
Dist.	$3.37^{\circ}$	$1.20^{\circ}$	-1	$16.37^{\circ}$	$4.70^{\circ}$	$5.92^{\circ}$

Table 2. Minimum inter-TC distance ( ∘ ) within each region from 2000 to 2023. A value of –1 indicates no twin TCs occurred.

Figure 3. Bias in track and intensity predictions. Solid lines correspond to forecast models:BaguanCyclone (ours), HRES (ECMWF), Vanilla Baguan, Pangu-Weather, and Graphcast. As ERA5 (the horizontal dashed lines) represents reanalysis data rather than predictive forecasts, its 6-hour intrinsic error is depicted as a constant baseline. The gray histograms indicate the sample count of the validation set. A three-panel figure…

Figure 4. Forecast evaluation for tropical cyclone Beryl: (a) track error, (b) max wind error, and (c) observation vs. prediction. Black line and gray shading indicate landfall and re-emergence; dashed line denotes mean error. A three-panel figure…

Figure 5. Forecast evaluation for tropical cyclone Pulasan, following the same format as Fig. 4 . A three-panel figure…

Figure 6. Forecast evaluation for tropical cyclone Trami and Kong-rey, following the same format as Fig. 4 . A three-panel figure…

Figure 7. Forecast evaluation for tropical cyclone Shanshan, following the same format as Fig. 4 . A three-panel figure…

Lead	WP			NA
Time	Vanilla	Prob.	Res.	Vanilla	Prob.	Res.
24h	60.9	53.0	60.7	53.8	48.6	49.8
48h	98.8	91.5	101.1	92.8	88.5	87.6
72h	157.1	147.9	158.4	112.8	109.6	111.4

Table 3. A comparative analysis of probabilistic and residual forecasting in the tracking refinement module in WP and NA. The error is measured in km. Lower value is better.

Lead	12h	24h	36h	48h	60h	72h
wind (m/s)	+0.89	+1.50	+2.40	+2.88	+3.74	+0.15
pres (hpa)	+1.96	+0.82	+1.87	+0.81	+1.81	-0.55

Table 4. Performance comparison of coupled vs. separate SI models. Values indicate % improvement; positive values favor the coupled model.

Figure 8. Average real-time forecasting performance during the 2025 WP typhoon season. Note that most baseline models (e.g., FuXi and Pangu-Weather) are not the latest versions.

Tropical Cyclone Intensity Forecasting

See all 14 papers

Can AI Forecast the Extremes? Evaluating Probabilistic Weather Models on Tropical Cyclones
Oct 15, 2026Nathan Chalumeau, Xavier Renard, Remi Meynadier +3Tropical Cyclone Intensity Forecasting Weather Forecasting
INRIA, Paris, France · AXA, Paris, France · Sorbonne Université, CNRS, LIP6, F-75005 Paris, France +1
3DTCR: Vortex-following generative reconstruction improves tropical cyclone intensity prediction from global AI weather forecasts
Mar 13, 2026Jun Liu, Xiaohui Zhong, Kai Zheng +5Tropical Cyclone Intensity Forecasting Weather Forecasting
Artificial Intelligence Innovation and Incubation Institute, Fudan University, Shanghai, 200433, China. · Department of Atmospheric and Oceanic Sciences, Institute of Atmospheric Sciences, Fudan University, Shanghai, 200438, China. · Shanghai Academy of Artificial Intelligence for Science, Shanghai, 200232, China.
Tianmu-TC: Physics-constraints Generative Artificial Intelligence for Global Tropical Cyclone Forecasting
Aug 19, 2026Shiqi Zhang, Pan Mu, Cheng Huang +6Tropical Cyclone Intensity Forecasting Weather Forecasting
College of Computer Science and Technology, Zhejiang University of Technology, Hangzhou & 310023, China. · School of Control Science and Engineering, Shandong University, Jinan & 250061, China. · School of Computer Science and Engineering, Tianjin University of Technology, Tianjin & 300384, China. +1

Oct 15, 2026J/K move · Enter open · S save

Can AI Forecast the Extremes? Evaluating Probabilistic Weather Models on Tropical Cyclones

Nathan Chalumeau, Xavier Renard, Remi Meynadier, Hugo Rakotoarimanga+2

INRIA, Paris, France · AXA, Paris, France · Sorbonne Université, CNRS, LIP6, F-75005 Paris, France+1

Tropical Cyclone Intensity Forecasting Weather Forecasting Extreme Weather Forecasting Probabilistic Forecasting Forecasting Benchmarks

Open paper AskPDF arXiv

Weather prediction is undergoing an important shift, with data-driven AI models outperforming leading numerical weather prediction systems on average at a fraction of the computational cost. Yet, whether this success extends to extreme events remains contested. Earlier evaluations have exposed important shortcomings, but have largely focused on deterministic models and metrics, leaving the capabilities of the latest probabilistic models less understood. We conduct an extensive benchmark of tropical cyclone forecasting, comparing GenCast, FourCastNet 3, and FGN with ECMWF's operational physics-based ensemble, IFS-ENS. We develop a suite of probabilistic metrics and diagnostic analyses to assess cyclone tracking, intensity biases, uncertainty, and the wind-pressure relationship that characterizes storm structure. We find that, despite remaining limitations in intensity and ensemble spread, leading AI models substantially outperform IFS-ENS in cyclone tracking: FGN achieves better probabilistic location scores for 90% of evaluated storms in 2025. Further analyses reveal model-specific failure modes and suggest a lingering influence of pretraining data, pointing to opportunities for targeted improvement. Nonetheless, these results suggest that AI models are capable of capturing physical relationships that remain predictive well beyond typical weather conditions, motivating further investigation into how these capabilities emerge and how to strengthen them.

Abstract

Tropical Cyclone Intensity Forecasting

Explore similar work

Can AI Forecast the Extremes? Evaluating Probabilistic Weather Models on Tropical Cyclones

3DTCR: Vortex-following generative reconstruction improves tropical cyclone intensity prediction from global AI weather forecasts

Tianmu-TC: Physics-constraints Generative Artificial Intelligence for Global Tropical Cyclone Forecasting