Tornadoes are severe convective weather exhibiting localized and sudden occurrences. Weather radar is widely regarded as the most effective tool for monitoring tornadoes and issuing early warnings. Dual-polarization updating has significantly improved tornado prediction and forecasting abilities. This article proposes an innovative tornado detection algorithm based on XGBoost which is suitable for dual-polarization radar data, was upgraded and has been used in China since 2019, and has been applied in the Tornado Key Open Laboratory of the China Meteorological Administration. The characteristics associated with the velocity attributes, reflectivity, velocity spectrum width, differential reflectivity, and correlation coefficient are used in the algorithm to achieve real-time tornado detection. Experimental evaluations have demonstrated that the proposed algorithm significantly improves tornado detection rates and leading times. Compared with the traditional TDA-RF algorithm based on Doppler weather radar data, the TDA-XGB algorithm introduces dual polarization parameters (such as differential reflectivity and the correlation coefficient), which effectively improves tornado identification performance. In addition, the TDA-XGB algorithm combines artificial intelligence-assisted learning to optimize the traditional algorithm based on the tornado vortex signature (TVS) and tornado debris signature (TDS), further improving the detection effect. Furthermore, the algorithm provides classification probabilities in the genesis and evolution of tornadoes, thereby supporting forecasters in their efforts to anticipate and issue timely tornado warnings.

Tornado is a localized, small-scale severe convective weather phenomenon characterized by extreme destructiveness. Tornado detecting and warning mainly rely on Doppler weather radar, which identifies and tracks tornadoes by recognizing the tornado vortex signature and supercells in radar data. Artificial intelligence technology has been applied to tornado recognition in recent years. However, existing monitoring methods, especially those using unsupervised learning algorithms, still have limited recognition accuracy and timely warning, and usually struggle to strike a balance between detection accuracy and false alarm rate. A novel tornado detection algorithm TDA-DARKNet has been proposed to address the aforementioned issues. The algorithm integrates a dual attention mechanism, dense residual connections, and Kolmogorov–Arnold network (KAN). A tornado dataset suitable for deep learning has been formed, which utilizes features including radial velocity, reflectivity, velocity spectrum width, differential reflectivity, and correlation coefficient in radar data. The TDA-DARKNet algorithm was trained and tested using the tornado dataset, and evaluated in tornado cases. The experimental results show that TDA-DARKNet improves the detection probability and extends the lead time to a maximum of 42 min in strong tornado situations, while achieving 97.11% accuracy, 95.08% precision, indicating strong overall identification performance. In addition, by directly leveraging radar-based data for tornado identification, the algorithm eliminates the need for manual feature engineering, simplifies data processing, reduces complexity, and further enhances detection effectiveness.