AI summaryⓘ
The authors looked at ways to fill in missing information about tropical cyclones, specifically the radius of maximum winds (Rmax), which is important for predicting coastal hazards. They tested different computer models, including ones that understand sequences over time, and found models using time data worked better even with less training data. Adding information about the radius of 34-knot winds (R34) made all models more accurate. They also found that using synthetic data to help train the models did not improve results, possibly because the synthetic data didn't match real-world conditions well. Overall, the study shows that using time-based deep learning and physics-related inputs can help improve predictions from incomplete storm data.
Radius of Maximum Winds (Rmax)Tropical Cyclone (TC)Joint Probability Method1D Convolutional Neural NetworksLong Short-Term Memory (LSTM)Transfer LearningSynthetic DatasetsIBTrACSRadius of 34-knot Winds (R34)Temporal Modeling
Authors
Swastik Agrawal, Nishkal Hundia, Ziyue Liu, Michelle Bensi
Abstract
Probabilistic coastal hazard assessments require accurate characterization of tropical cyclone (TC) parameters, yet datasets often contain missing records for the radius of maximum winds (Rmax), a key variable in Joint Probability Method analyses. This study evaluates data-driven approaches for Rmax imputation, including one-dimensional Convolutional Neural Networks (1DCNNs), Long Short-Term Memory (LSTM) networks, and conventional machine learning models. We examine physics-informed input augmentation, temporal modeling, and transfer learning using synthetic RAFT and STORM datasets for pre-training and observational IBTrACS data for fine-tuning. Including the radius of 34-knot winds (R34) substantially improves performance across all model types. Temporal models achieve higher average correlations than non-temporal models despite using approximately an order of magnitude fewer samples, indicating better preservation of relative Rmax variability across storms. This advantage is more pronounced when R34 is unavailable, suggesting temporal information can partially compensate for missing storm-size predictors. Transfer learning does not improve performance, likely because synthetic datasets have lower and less variable Rmax distributions than IBTrACS. These findings demonstrate the potential of temporal deep learning for reconstructing incomplete TC records and highlight the importance of physics-informed inputs, observational data availability, and distributional consistency in coastal hazard assessment.