Beyond the Backbone: A Quantitative Review of Deep-Learning Architectures for Tropical Cyclone Track Forecasting
Accurate forecasting of tropical cyclone (TC) tracks is critical for disaster preparedness and risk mitigation. While traditional numerical weather prediction (NWP) systems have long served as the backbone of operational forecasting, they face limitations in computational cost and sensitivity to ini...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-08-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/17/15/2675 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849406429315727360 |
|---|---|
| author | He Huang Difei Deng Liang Hu Yawen Chen Nan Sun |
| author_facet | He Huang Difei Deng Liang Hu Yawen Chen Nan Sun |
| author_sort | He Huang |
| collection | DOAJ |
| description | Accurate forecasting of tropical cyclone (TC) tracks is critical for disaster preparedness and risk mitigation. While traditional numerical weather prediction (NWP) systems have long served as the backbone of operational forecasting, they face limitations in computational cost and sensitivity to initial conditions. In recent years, deep learning (DL) has emerged as a promising alternative, offering data-driven modeling capabilities for capturing nonlinear spatiotemporal patterns. This paper presents a comprehensive review of DL-based approaches for TC track forecasting. We categorize all DL-based TC tracking models according to the architecture, including recurrent neural networks (RNNs), convolutional neural networks (CNNs), Transformers, graph neural networks (GNNs), generative models, and Fourier-based operators. To enable rigorous performance comparison, we introduce a Unified Geodesic Distance Error (UGDE) metric that standardizes evaluation across diverse studies and lead times. Based on this metric, we conduct a critical comparison of state-of-the-art models and identify key insights into their relative strengths, limitations, and suitable application scenarios. Building on this framework, we conduct a critical cross-model analysis that reveals key trends, performance disparities, and architectural tradeoffs. Our analysis also highlights several persistent challenges, such as long-term forecast degradation, limited physical integration, and generalization to extreme events, pointing toward future directions for developing more robust and operationally viable DL models for TC track forecasting. To support reproducibility and facilitate standardized evaluation, we release an open-source UGDE conversion tool on GitHub. |
| format | Article |
| id | doaj-art-5ec9d10c8fca450daf41ddbf0df69949 |
| institution | Kabale University |
| issn | 2072-4292 |
| language | English |
| publishDate | 2025-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj-art-5ec9d10c8fca450daf41ddbf0df699492025-08-20T03:36:22ZengMDPI AGRemote Sensing2072-42922025-08-011715267510.3390/rs17152675Beyond the Backbone: A Quantitative Review of Deep-Learning Architectures for Tropical Cyclone Track ForecastingHe Huang0Difei Deng1Liang Hu2Yawen Chen3Nan Sun4School of Systems and Computing, University of New South Wales at Canberra, Canberra 2612, AustraliaSchool of Science, University of New South Wales at Canberra, Canberra 2612, AustraliaGeoscience Australia, Canberra 2609, AustraliaSchool of Systems and Computing, University of New South Wales at Canberra, Canberra 2612, AustraliaSchool of Systems and Computing, University of New South Wales at Canberra, Canberra 2612, AustraliaAccurate forecasting of tropical cyclone (TC) tracks is critical for disaster preparedness and risk mitigation. While traditional numerical weather prediction (NWP) systems have long served as the backbone of operational forecasting, they face limitations in computational cost and sensitivity to initial conditions. In recent years, deep learning (DL) has emerged as a promising alternative, offering data-driven modeling capabilities for capturing nonlinear spatiotemporal patterns. This paper presents a comprehensive review of DL-based approaches for TC track forecasting. We categorize all DL-based TC tracking models according to the architecture, including recurrent neural networks (RNNs), convolutional neural networks (CNNs), Transformers, graph neural networks (GNNs), generative models, and Fourier-based operators. To enable rigorous performance comparison, we introduce a Unified Geodesic Distance Error (UGDE) metric that standardizes evaluation across diverse studies and lead times. Based on this metric, we conduct a critical comparison of state-of-the-art models and identify key insights into their relative strengths, limitations, and suitable application scenarios. Building on this framework, we conduct a critical cross-model analysis that reveals key trends, performance disparities, and architectural tradeoffs. Our analysis also highlights several persistent challenges, such as long-term forecast degradation, limited physical integration, and generalization to extreme events, pointing toward future directions for developing more robust and operationally viable DL models for TC track forecasting. To support reproducibility and facilitate standardized evaluation, we release an open-source UGDE conversion tool on GitHub.https://www.mdpi.com/2072-4292/17/15/2675tropical cyclone forecastingdeep learningartificial intelligencetrack predictiontropical cyclone forecasting review |
| spellingShingle | He Huang Difei Deng Liang Hu Yawen Chen Nan Sun Beyond the Backbone: A Quantitative Review of Deep-Learning Architectures for Tropical Cyclone Track Forecasting Remote Sensing tropical cyclone forecasting deep learning artificial intelligence track prediction tropical cyclone forecasting review |
| title | Beyond the Backbone: A Quantitative Review of Deep-Learning Architectures for Tropical Cyclone Track Forecasting |
| title_full | Beyond the Backbone: A Quantitative Review of Deep-Learning Architectures for Tropical Cyclone Track Forecasting |
| title_fullStr | Beyond the Backbone: A Quantitative Review of Deep-Learning Architectures for Tropical Cyclone Track Forecasting |
| title_full_unstemmed | Beyond the Backbone: A Quantitative Review of Deep-Learning Architectures for Tropical Cyclone Track Forecasting |
| title_short | Beyond the Backbone: A Quantitative Review of Deep-Learning Architectures for Tropical Cyclone Track Forecasting |
| title_sort | beyond the backbone a quantitative review of deep learning architectures for tropical cyclone track forecasting |
| topic | tropical cyclone forecasting deep learning artificial intelligence track prediction tropical cyclone forecasting review |
| url | https://www.mdpi.com/2072-4292/17/15/2675 |
| work_keys_str_mv | AT hehuang beyondthebackboneaquantitativereviewofdeeplearningarchitecturesfortropicalcyclonetrackforecasting AT difeideng beyondthebackboneaquantitativereviewofdeeplearningarchitecturesfortropicalcyclonetrackforecasting AT lianghu beyondthebackboneaquantitativereviewofdeeplearningarchitecturesfortropicalcyclonetrackforecasting AT yawenchen beyondthebackboneaquantitativereviewofdeeplearningarchitecturesfortropicalcyclonetrackforecasting AT nansun beyondthebackboneaquantitativereviewofdeeplearningarchitecturesfortropicalcyclonetrackforecasting |