A Novel Hybrid Approach for Tornado Prediction in the United States: Kalman-Convolutional BiLSTM with Multi-Head Attention
Jiawei Zhou
TL;DR
This work tackles the challenge of accurate tornado prediction amid high false alarm rates by leveraging the Seamless Hybrid Scan Reflectivity (SHSR) dataset from MRMS. It introduces a novel Kalman Filtering–Convolutional BiLSTM–Multi-Head Attention architecture that jointly models dynamic state estimation, spatial features, and temporal dependencies to improve warning reliability. Across comparisons with KNN, LightGBM, SVM, and recurrent architectures, the proposed model delivers superior precision, recall, F1-score, and overall accuracy, indicating strong potential for reducing false alarms while maintaining timely detections. The approach advances data-driven tornado forecasting with robust state estimation and attention-based feature weighting, and points to future integration with larger language models for interpretability and decision support in meteorology.
Abstract
Tornadoes are among the most intense atmospheric vortex phenomena and pose significant challenges for detection and forecasting. Conventional methods, which heavily depend on ground-based observations and radar data, are limited by issues such as decreased accuracy over greater distances and a high rate of false positives. To address these challenges, this study utilizes the Seamless Hybrid Scan Reflectivity (SHSR) dataset from the Multi-Radar Multi-Sensor (MRMS) system, which integrates data from multiple radar sources to enhance accuracy. A novel hybrid model, the Kalman-Convolutional BiLSTM with Multi-Head Attention, is introduced to improve dynamic state estimation and capture both spatial and temporal dependencies within the data. This model demonstrates superior performance in precision, recall, F1-Score, and accuracy compared to methods such as K-Nearest Neighbors (KNN) and LightGBM. The results highlight the considerable potential of advanced machine learning techniques to improve tornado prediction and reduce false alarm rates. Future research will focus on expanding datasets, exploring innovative model architectures, and incorporating large language models (LLMs) to provide deeper insights. This research introduces a novel model for tornado prediction, offering a robust framework for enhancing forecasting accuracy and public safety.